HU215243B - Method for production of conjugates - Google Patents

Abstract

Conjugates useful for treating gastrointestinal tumours such as colorectal tumours comprises an antibody such as C242 antibody and a toxin such as ricin A. The antibody and toxin may be coupled by a linker. Pharmaceutical compositions containing the conjugates and processes for their preparation are also inculded as well as methods of treating gastrointestinal tumours.

Description

The present invention relates to conjugates, including immunotoxins for use in the treatment of cancer, and to pharmaceutical compositions containing them.

The ricin and ricin-like molecules (such as abrin, modecin and viscumin) are known compounds produced by plant cells that have cytotoxic effects. Toxins of this type consist of two polypeptide chains linked by a disulfide bridge. One of the polypeptide chains (chain A) is responsible for the cytotoxic effect of the toxin molecule, while the other polypeptide chain (chain B) allows the toxin molecule to attach to the cell surface.

The toxicity of ricin-like molecules occurs through three phases.

(a) the toxin binds to the cell surface by interaction of galactose binding sites on the B chain with glycoproteins or glycolipids on the cell surface;

(b) at least chain A penetrates into the cellular fluid; and (c) the A chain breaks down the activity of the 60S subunits of the ribosomes to inhibit protein synthesis.

According to some assumptions, the primary secondary function of the B chain, in addition to attaching the toxin molecule to the cell surface (primary function), is to facilitate the uptake of toxins by cells. Experience has shown that the separated "A" and "B" chains are essentially non-toxic; in fact, in the absence of the B chain, the A chain cannot bind to the cell surface and penetrate the cellular fluid, whereas the B chain has no cytotoxic effect.

As previously reported, ricin and ricin-like molecules are produced by plants. The castor is produced by Ricinus communis (castor plant). To date, the plant is the first to produce a precursor polypeptide called preproricin, which contains a leader sequence, A chain, linker sequence and B chain. During transport to the plant cells from preproricin, the leader sequence is cleaved to produce proricin. The proricin is then formed by the formation of the disulfide bridge between the "A" and "B" chains and the cleavage of the linker to form ricin.

Previously, attempts have been made to exploit the toxicity of the "A" chain of toxins, including ricin, in cancer therapy by replacing the non-target cell-associated "B" chain with a carrier that preferentially binds to normal tumor cells. On this basis, various conjugates were formed consisting of ricin or the "A" chain of ricin and tumor specific antibody. However, such conjugates, immunoconjugates or immunotoxins have several disadvantages.

One of the problems with the known conjugates is due to the structural properties of the "A" chain of natural ricin. Indeed, the synthesis of ricin in plant cells undergoes N-glycosylation and the resulting sugar moieties may interact with the cell surface in a non-specific manner.

The low selectivity and the lack of selectivity of plant ricin-containing immunoconjugates is also due to (a) non-specific interaction of the B-galactose binding sites on the cell surface with whole ricin-containing immunoconjugates; or (b) in the case of immunoconjugates containing plant castor A chain, the glucan side chains expressed on plant castor A chain bind to galactose mannose receptors on the cell surface.

Further problems arise from the fact that, as already mentioned, the uptake of toxins by cells is believed to be facilitated by the "B" chain. Thus, although conjugates containing no "B" chain, i.e. conjugates of A chain and antibody, are more specific than natural toxins, they do not possess the potency of a natural toxin [Moolten F. L. et al., Immun. Rev., 62.47-72 (1982)]. This decrease in activity is probably due to the lower uptake of toxin by cells in the absence of the B chain. Only a relatively small proportion of the antibodies tested are those that adequately promote toxin intimalization, resulting in the formation of immunotoxins of reduced potency or ineffectiveness in most cases.

To date, several conjugates have been prepared that may be effective against tumors in various tissues. For example, WO 85/003508 discloses conjugates in which antibodies specific for respiratory tumors are linked via the insertion of a spacer molecule of ricin or diphtheria toxin A chain. Anti Cancer Drug Design 1, 179-188 (1986) discloses immunotoxins in which the toxin is linked via three different types of linker (one containing a reducing disulfide bridge, a protected disulfide bridge, and a non-reducing sulfide bridge). -LOND-Fib75 antibody. This antibody directs the immunotoxin to breast tissue. European Patent 031999 discloses protein hybrids in which certain known anti-tumor antibodies bind to the A 'chain of diphtheria toxin.

Antibodies that recognize gastrointestinal tumor cells have also become known; however, conjugates containing such antibodies bind unacceptably to normal tissues and / or have low potency. An example is the conjugate described in the Journal of Biological Response Modifiers, 7, 559-567 (1988), which contains an anti-transferrin receptor antibody linked to the "A" chain of ricin. Although the transferrin receptor is present in very high levels in some tumors, such as colon and pancreatic adenocarcinomas, it is a disadvantage that this receptor (a cell surface protein) is also found in many other cells of the human body. Thus, the conjugate disclosed in the cited publication is not specific enough to be used for therapeutic purposes and can severely damage normal cells. Col. 2

Until now, conjugates with real therapeutic value have not been developed using known antibodies recognizing rectal tumor cells. Therapeutic value conjugates for the treatment of colorectal tumors are in great need.

In general, although several conjugates have been prepared so far, conjugates with improved properties are still needed to eliminate or reduce one or more disadvantages associated with known conjugates. There is also a need for conjugates that selectively target gastrointestinal tumors.

The present invention provides conjugates comprising a toxin component and a target cell binding component that is selectively bound to gastric and intestinal tumor cells.

Preferred conjugates of the present invention are those in which the target cell-binding components (and consequently also the conjugates) selectively bind to colorectal and pancreatic tumor cells, particularly colorectal tumor cells.

The target cell binding components may be, for example, antibodies, antibody fragments, or derivatives thereof.

By selectivity for binding to the target cell-binding component, it is understood that this component preferentially binds to gastric and intestinal tumor cells, preferably colorectal tumor cells, relative to normal cells. Accordingly, the target cell-binding component binds to gastric and intestinal tumor cells, while its ability to bind to normal cells is minimal or not at all capable of binding to normal cells. Preferred representatives of the target cell binding components are those which bind to gastric and intestinal tumor cells with substantially the same selectivity as the antibody to C242. The selectivity of binding of C242 antibody to gastric and intestinal tumor cells (e.g., colorectal tumor cells) is characterized by immunohistochemical data to be provided. The target cell-binding component, such as the C242 antibody, is primarily associated with antigens present in gastrointestinal tumors which are not or only very poorly expressed in normal cells. Examples of target cell binding components include the C242 antibody and other binding components having substantially the same cell binding characteristics.

As mentioned above, the conjugates of the present invention are capable of killing gastrointestinal tumor cells. Accordingly, conjugates (or "immunotoxins") are capable of delivering a toxin component to a tumor cell so that it can exert its cytotoxic effect on the tumor cell. In other words, the conjugate (via the tumor cell binding component, which binds to the target cell) is able to bind to the tumor cells and then the toxin component can penetrate the cell (i.e., the toxin component can "intimalize"). The general requirements for highly potent conjugates are:

1. The target cell binding component is capable of binding to the surface antigen of a tumor cell.

2. The cell surface antigen should be present in tumor cells in high copy number (e.g., at least 10,000 cells per cell).

3. The antigen should not be expressed in high copy numbers in normal cells.

4. The antibody / antigen complex should be effectively intimalized so that the toxin component can exert its intracellular cytotoxic activity.

5. The conjugate is preferably formulated in a form that is stable enough to remain stable in the bloodstream to allow the toxin components to reach the tumor cell, yet is easily cleaved to release the toxin component that has entered the cell.

Antibody C242 is an IgG class of murine antibody produced by culturing spleen cells from mice immunized with human colon adenocarcinoma cell line and a hybridoma cell line formed by Sp 2/0 murine myeloma cell line. The 242: 11 hybridoma cell line producing C242: II antibody (simplified term C242 antibody) was deposited on January 26, 1990 in the European Collection of Animated Cell Cultures (ECACC) (PHLS Center for Applied Microbiology and Research). , Porton Down, Salisbury, Wiltshire, Great Britain) under the number 90012601.

Information on the potential cancer use of the C242 antibody is available in European Journal of Surgical Oncology, 15, 367-370 (1989), which discloses the use of two forms of monoclonal antibodies, C242 and C215, in the treatment of transplanted rodent mammary tumors. The antitumor activity of the preparations tested there is due to the presence of the radioactive isotope; treatment with the compositions has the well-known drawbacks of radiotherapy. This publication makes no reference to the combination of antibodies with toxins or the possibility of selective treatment of gastrointestinal tumors.

The term "target cell binding component having substantially the same cell binding properties as the antibody to C242" refers to the target cell binding components having substantially the same immunological specificity as the antibody produced by the cell line deposited under ECACC 90012601, i.e., having the same antigen or is a competitor of C242 for binding to the site.

By "target cell binding component" is meant, in addition to whole antibodies, antibody fragments, such as antibody fragments of C242. Found3

Antibody fragments which retain the binding and specificity of the non-fragmented immunoglobulin may be used for the purpose of this disclosure. Such fragments are formed, for example, when the antibody is degraded by a proteolytic enzyme, such as pepsin. Examples of fragments include the F (ab ') and F (ab') ₂ fragments.

The target cell binding components may also be engineered by genetic engineering techniques. Accordingly, the term "target cell binding component" is intended to include, in addition to the above, components engineered by genetic engineering (i.e., antibodies and antibody fragments generated by genetic engineering). Preferred representatives of these are genetic engineering C242 antibody and fragments thereof.

The production of antibody fragments, including primarily humanized antibody fragments, is described, inter alia, in Carter et al., Biotechnology 10, February 1992.

In addition, the target cell binding components may be derivatives of antibodies and antibody fragments. These include humanized forms of non-human derived antibodies or antibody fragments, such as humanized forms of murine antibodies.

The C242 antibody is directed against a tumor associated antigen, hereinafter referred to as CA-242 antigen. Accordingly, antibodies or antibody fragments capable of binding to the CA-242 antigen are preferred representatives of the target cell binding components. One representative of such antibodies is the C242 antibody itself. Tumor-associated CA-242 antigen is expressed in the majority of colorectal tumors but is not or only poorly expressed in normal colon tissue.

During the aforementioned genetic operations, cDNAs within the variable region of the light and heavy chains of the C242: II monoclonal antibody were cloned. This procedure is described below. However, before describing the genetic operations, reference will be made to Figure 17 (showing the general structure of a class G antibody, i.e., immunoglobulin), which briefly reviews the basic structural units of immunoglobulin.

The immunoglobulin of the structure shown in Figure 17 consists of two identical light L polypeptide chains and two identical heavy H polypeptide chains; the four chains are linked together by disulfide bonds and various non-covalent bonding forces in a symmetric "Y" configuration. At the end of each heavy chain is a variable region (V _H ) followed by several constant regions (C _H ); while light chains have a variable range (V _L ) at one end and a constant range (C _L ) at the other end. There are two types of light chains, the kappa type and the lambda type. The variable region of the light chain is located along the variable region of the heavy chain, while the constant region of the light chain is located along the first constant region of the heavy chain; the variable regions of the light and heavy chains form the antigen binding site.

The variable structure of the variable region of the light and heavy chains is the same: each contains three hypervariable regions (complementarity determining regions or CDRs) located between four framework regions (FRs). The CDRs of each variable region are closely held together by the framework regions; the specificity of the antibody is provided by the combination of the two CDR series.

For the cloning of cDNAs in the variable regions of the light and heavy chains of the C242: II antibody, a cDNA phage library was first generated by methods known from the C242: II hybridoma. Hybridization probes were then prepared from mouse genomic DNA using polymerase chain reaction (PCR), using appropriate primers, covering the constant portions of the heavy and light chains (kappa). The probes thus obtained were screened for the presence of cDNA clones containing DNA sequences encoding the heavy chain and kappa chain of antibody C242: II. Positively hybridizing phage clones were propagated and cDNA was excised as plasmids. The latter were analyzed by restriction enzyme mapping and then the plasmids containing the expected size insert were sequenced. To determine the CDR regions, the amino acid sequences encoded by the sequences of the insert portions were compared with those of the corresponding portions of the known murine kappa and heavy chains; baseline CDR regions were considered based on the publication of Kábát, E.A., et al., Sequence of Proteins of Immunological Interest, 4th U.S. Pat. Department of Health and Human Services, National Institutes of Health (1987)]. In our study, the amino acid sequence of the corresponding chains was as follows (see also SEQ ID NO: 21 and SEQ ID NO: 21 and 22): Light chain:

CDR1 (SEQIDNo. 1): ArgSerSerLysSerLeuLeuHisSerAsnGlyAsnThrTyrLeu Tyr

CDR2 (SEQ ID No. 2): ArgMetSerAsnLeuValSer CDR3 (SEQ ID No. 3): LeuGlnHisLeuGluTyrProPheThr

Heavy chain:

CDR1 (SEQ ID No. 4): TyrThrGlyMetAsn CDR2 (SEQ ID No. 5): TrpIleAspThrThrThrGlyGluProThrTyrAlaGluAspPheLysGly

CDR3 (SEQ IDNo. 6): ArgGlyProTyrAsnTrpTyrPheAspVal

Given the CDR amino acid and / or DNA sequences listed, the CDR regions of antibody C242 can be easily incorporated into cloned variable regions of another antibody or antibody fragment by known site-directed mutagenesis. In this sequence known as "CDR grafting" (described, for example, in EP-A-239400), the CDR coding sequences of the recipient antibody or antibody fragment at the DNA level are replaced with the CDR coding sequences of antibody C242: II. As a result, the procedure has the same specificity as the C242: II monoclonal antibody

An antibody or antibody fragment is obtained. Accordingly, the target cell-binding component may also be a component comprising at least one antibody or antibody fragment comprising a CDR as defined above (or substantially identical thereto), wherein the number of CDRs in the antibody or antibody fragment and arrangement results in a targeted cell binding component having substantially the same cell binding characteristics as the C242 antibody. Preferably, the target cell binding component comprises all of the CDRs listed above (or substantially the same) (i.e., light chain CDR1, CDR2 and CDR3, and heavy chain CDR1, CDR2, and CDR3). For example, CDRs that are substantially identical to the CDRs listed above may be extended sequences, of which the following are exemplified:

Light chain:

CDR1 (SEQ ID No. 7): ArgSerSerLysSerLeuLeuHisSerAsnGlyAsnThrTyrLeuTyrTrpPhe

CDR2 (SEQ IDNo. 8): IleTyrArgMetSerAsnLeuValSerGlyVal

CDR3 (SEQ ID No. 9): LeuGlnHisLeuGluTyrProPheThrPheGly

Heavy chain:

CDR1 (SEQ ID No. 10): PheThrTyrThrGlyMetAsn CDR2 (SEQIDNo. 11): MetGlyTrpIleAspThrThrThrGlyGluProThrTyrAlaGluAspPheLysGly Arglle

CDR3 (SEQIDNo. 12): AlaArgArgGlyProTyrAsnT rpTyrPhe Asp V alTrpGly

In the CDR grafting operation, the antibodies or antibody fragments are preferably produced by grafting the C242 CDRs into a frame that is close to C242 in terms of both homology and CDR size.

Preferably, the present invention provides conjugates comprising a toxin component and a target cell binding component comprising a C242 antibody or a target cell binding component having substantially the same cell binding properties as the C242 antibody.

It should be noted that the conjugates prepared according to the invention need not necessarily be products constructed from a toxin molecule and an antibody molecule. For example, conjugates may contain more than one toxin molecule per antibody molecule.

The toxin component of the conjugates will usually be a toxin component having a cytotoxic effect which, upon intimalization, can kill the target cells.

The target cell-binding component of the conjugates may typically be a target cell-binding component capable of recognizing a specific antigenic determinant or a specific target cell.

The toxin component and the target cell binding component may be linked directly or indirectly. The toxin component and the target cell binding component are usually coupled such that the geometry of the conjugate formed allows the target cell binding component to bind to the target cell. Preferably, the toxin component and the target cell binding component are coupled to form a conjugate that is stable in the extracellular environment but unstable in the extracellular environment, whereby the toxin component and the target cell binding component remain linked outside the target cell (penetration into the target cell). however, the toxin component is released from the conjugate after intimalization). Accordingly, the conjugate preferably comprises an intracellular cleavable / extracellular site.

Conjugates containing the toxin component and the target cell-binding component directly linked to one another include, for example, conjugates in which the toxin component and the target cell-binding component are coupled by the reaction of a thio group on the toxin component with a thio sulfide on the target cell-binding component.

The target cell binding component may also comprise a region that recognizes the toxin component and thereby binds the toxin component to the target cell binding component. For example, the target cell binding component comprises an antibody or antibody fragment that binds to the toxin component.

The target cell binding component and the toxin component may form a single polypeptide, preferably comprising an intracellular cleavage site such that the toxin component is released from the molecule within the cell. In such polypeptides, the toxin component and the target cell-binding component are linked via a polypeptide coupling agent.

Conjugates in which the toxin component is indirectly linked to the target cell-binding component include, for example, products in which the toxin component is linked to the target cell-binding component via a bi-functional coupling agent, preferably a hetero-bifunctional coupling agent.

Coupling agents include, for example, polypeptide type coupling agents (such as those disclosed in WO 85/003508) or coupling agents containing reactive chemical groups (such as those disclosed in European Patent Application 169,119).

Preferably, the toxin component and the target cell binding component are linked by disulfide bond or thioether bond. Preferably, the toxin component and the target cell-binding component are coupled to a coupling agent that creates a disulfide or thioether bond that can be cleaved intracellularly between the two components. Such coupling agents and their use in the preparation of immunotoxins are described, for example, in European Patent Application No. 169,111; Methods in Enzymology, 112, 207-225 (1985) (Cumber, A.J., Forrester, J.A., Foxweel, J.M.,

HU 215 243 Β

Ross, W. C. J., Thorpe, P. E.: Preparation of antibodytoxin conjugates); and Vogel, Immunoconjugates: Antibody Conjugates in Radioimaging and Therapy of Cancer, 28-55. (Wawrzynczak, E.J. and Thorpe, P.E.: Methods for Preparing Immunotoxins: Effects of Linkage on Activity and Stability).

One class of coupling agent is a substance which is capable of forming a disulfide bond with a thio group on one of the toxin component and the target cell binding component and an amide bond with an amino group on the other of the toxin component and the target cell binding component.

These coupling agents contain a moiety of the formula -S-X-CO- wherein X is a spacer group.

X is, for example, one or two straight-chain alkyl groups optionally substituted with one or two C 1 -C 6 alkyl, phenyl and / or (C 1 -C 4 alkyl) phenyl, or optionally substituted on the alkyl moiety with one or two C 1 -C 6 alkyl, phenyl and and / or (C 1 -C 4 alkyl) phenyl optionally substituted with (C 1 -C 4 alkyl) phenyl and optionally substituted on the benzene ring with, for example, halogen, C 1 -C 4 alkyl or C 1 -C 4 alkoxy.

Preferably X is an alkylene group as defined above.

For example, X may be a C 1 -C 6 alkylene chain optionally substituted with one or two C 1 -C 6 alkyl and / or phenyl groups. X may be, for example, methylene, ethylene, propylene or butylene, to which may be optionally attached one or two methyl, ethyl, propyl and / or butyl groups. Preferably, the alkylene groups substituted by methyl, ethyl, propyl or butyl.

Preferably X is ethylene or methylene to which one or two of the abovementioned substituents may be attached.

X preferably represents unsubstituted or substituted ethylene by one or two methyl groups, particularly preferably ethylene substituted by one methyl group.

Preferred representatives of the conjugates of the present invention are

Compounds of Formula A-Si-S ₂ -X-CO-NH-B wherein A is a toxin component and a target cell-binding component, and

B is the other of the toxin component and the target cell binding component, wherein one or more toxin components may be attached to the target cell binding component,

Sj is the sulfur atom in moiety A, NH is in moiety B, and the -S ₂ -X-CO- moiety is the linker moiety, wherein

S ₂ is a sulfur atom, and X are as defined above.

In the above compounds, the sulfur atom S1 is derived from the thio moiety on the A moiety.

It will be appreciated that coupling agents containing an asymmetric center may be formed and isolated in the form of optically active modifications and racemates. Our need for protection extends to the use of optically active and racemic coupling agents.

The optically active coupling agents may be formed and isolated by known organic chemical methods.

In the compounds of the above general formula, preferably A moiety represents the toxin component and B moiety represents the target cell binding component.

As stated above, the toxin component can be any component that has cytotoxic properties. Such components are, for example, polypeptides that inactivate ribosomes and thereby inhibit protein synthesis. Examples of the latter polypeptides include naturally occurring polypeptides and components thereof. When a given polypeptide contains a substantially non-toxic component in addition to the toxic component, only the toxic component of that polypeptide is preferably used as the toxin component. Thus, for example, the toxin component may be whole ricin, but preferably only the "A" chain of ricin is used as the toxin component. The toxin component may also be a portion or analog of natural polypeptides having cytotoxic properties. Polypeptides useful as toxin components and their preparation are described in detail in Stripe F. and Barbieri L. "Ribosome inactivating proteins up to date" (FEBS 3269) and references cited therein. Particularly preferred polypeptide toxin components are the "A" chain of ricin (hereinafter "ricin-A"), the constrictocin and the Shiga toxin. Also preferred as the toxin component are the following polypeptides: abrin, viscumin, modecin, volcensin, gelonin, sub-claw antiviral protein, saparin, lufin, trichosanthine, barley ribosome inactivating protein, diantins, biodin, momordin, tritin, dodecandrin, toxin, pseudomonas exotoxin A and Shiga analogue toxins VT1 and VT2.

Analogs of naturally occurring polypeptides useful in the present invention may be polypeptides that differ from the native polypeptide in one or more amino acid changes (deletion, insertion, replacement) that do not result in loss of cytotoxic activity. Preferred examples of such analogs are compounds in which the introduced modification (s) facilitate the attachment of the molecule to the target cell binding component.

When the toxin component is part or all of a natural polypeptide, the toxin component can be prepared by isolating the polypeptide from its natural source and, if desired, chemically or enzymatically modifying the isolated polypeptide to produce a polypeptide moiety. The polypeptides or portions thereof may also be produced by genetic manipulation. Such methods include recombinant DNA technology, wherein the DNA sequence encoding the desired polypeptide is usually inserted into a suitable vector or plasmid, transformed with the vector or plasmid, and then the host cells are conveniently surrounded.

By culturing between cultures, the DNA sequence is expressed to form the desired polypeptide.

Preferably, the toxin component may be ricin A, a portion thereof, or an analogue thereof. For example, ricin A can be isolated from the seeds of Ricinus communis (ricin plant); however, in a more preferred embodiment, the "A" chain of ricin or ricin is produced by genetic methods. Preferably, the toxin component is recombinant ricin A or recombinant ricin.

The use of ricin A produced by recombinant DNA technology is generally preferable to the use of ricin A isolated from nature because ricin A produced by recombinant DNA technology does not contain the B chain as a contaminant and the ricin A thus produced is not glycosylated. Due to the lack of non-discriminating B chain and the lack of sugar moieties capable of interacting specifically with cell surfaces, recombinant ricin A can be used to form conjugates that are more selective than the use of naturally occurring ricin A.

The target cell binding component of the conjugates of the invention may preferably be a component selectively linked to colorectal tumor cells.

Preferably, the target cell binding component is a C242 antibody or fragment thereof, most preferably a C242 antibody.

Preferably, the toxin component and the target cell binding component are linked via a -CO-CH ₂ -CH (CH ₃ ) -S- linker derivative of 3-mercaptobutyric acid. This group forms a covalent bond with the -NH group on the target cell binding component (such as the -NH group on the lysyl residue) and forms a disulfide bridge with the thio group on the toxin component, such as the thio group on the cysteine residue of ricin A.

Thus, the present invention preferably provides an immunotoxin constructed from C242 antibody and recombinant ricin A, wherein an amino group of the C242 antibody is linked via the linker -CO-CH ₂ -CH (CH ₃ ) -S- to the cysteine residue in ricin A. thio group.

The latter immunotoxins particularly preferred constructions in the C242 antibody and recombinant ricin A from A immunotoxin, wherein the lysyl residue in C242 antibody terminal amino group of links cysteine residues terminal thio group in the ricin of -CO-CH ₂ -CH (CH ₃ ) -S-.

Generally, each antibody unit is linked to at least one ricin A unit. However, an antibody unit may be linked via appropriate linker groups to more than one linker group (s) that bind to the ricin A moiety). Non-toxic molecules, such as cysteine molecules, can be attached to these additional linker groups as terminating molecules.

The immunotoxins of the present invention are useful in medicine for the treatment of gastrointestinal tumors, in particular colorectal and pancreatic tumors. For the said therapeutic purposes, the conjugates of the invention are administered in effective amounts in the form of pharmaceutical compositions to warm-blooded persons, including humans.

The required dose of conjugates will vary depending on several factors, such as the quality of the toxin component used, the target cell population, and the patient's medical history. Typically, the dose of conjugate may be from 0.1 to 1 mg / kg body weight.

The invention further relates to a process for the preparation of a pharmaceutical composition containing the above conjugates as an active ingredient. These pharmaceutical compositions are prepared by known pharmaceutical technology using conventional pharmaceutical carriers, diluents and / or other excipients.

The pharmaceutical compositions may take a variety of forms. The conjugates are preferably formulated into compositions for parenteral administration, preferably intravenous administration. Preferred compositions for parenteral administration include injectable dosage units containing the immunotoxin in the form of a solution, emulsion or suspension in a carrier or diluent for parenteral administration. Carriers or diluents include, for example, aqueous vehicles (such as water or saline) and non-aqueous vehicles (e.g., stabilized oils or liposomes). The compositions may also contain stability enhancing agents, buffering agents, surfactants (such as Tween) and the like. The amount of conjugate in the injectable preparation may vary within a relatively wide range, however, per dosage unit will usually contain from about 1 mg to about 10 mg of the active ingredient.

Antibody C242 is a substance selectively associated with colorectal tumor cells. Conjugates containing such antibodies have been shown to be selective, potent immunotoxins for colorectal tumor cells.

Surprisingly, we have found that the preferred conjugates described above, while being selective for colorectal tumor cells, have been shown to be highly effective. It has also been found that the rate of conjugate blood clearance, extracellular cleavage and extracellular degradation is relatively low, i.e. compounds exhibit increased plasma persistence. This has the advantage that the immunotoxin can interact with the target cells for a sufficient time before degradation or elimination.

The preparation of the conjugates is described in more detail below.

a) Conjugates in which the target cell binding component is directly attached to the toxin component can be prepared by reacting the toxin component with the target cell binding component.

In the case where the target cell binding component is coupled to the toxin component via disulfide bridges, it is possible to proceed by targeting the target cell binding component prior to the reaction.

EN 215 243 on and one of the toxin components, for example dithiothreitol.

b) Conjugates in which the target cell binding component is linked to the toxin component via a linker moiety can be prepared by reacting the target cell binding component with a coupling agent or the toxin component coupling agent with the target cell.

The derivatives required for the latter reactions are formed by reacting the appropriate component with the coupling agent, thereby coupling the coupling agent to that component. Typically, the target cell binding component is derivatized with a coupling agent and reacted with the toxin component.

For example, the coupling agent derivative of the target cell binding component or the toxin component may be prepared by dissolving the component in a suitable buffer (such as acetate, phosphate, or borate) and adjusting the pH of the reaction mixture to a level compatible with the coupling agent. The coupling agent is then added to the reaction mixture. If the coupling agent is insoluble in the reaction mixture, it is also possible to add a solution of the component to a solution of the coupling agent in a small amount of an organic solvent such as dimethylformamide or dimethylsulfoxide. After completion of the reaction, the resulting derivative can be isolated by known methods, such as gel filtration, dialysis or column chromatography. The resulting target cell-binding component (or toxin component), which is then derivatized with a coupling agent, is then reacted with another component of the conjugate.

The coupling agents are selected according to the nature of the reactive groups (e.g., thio, carboxyl or amino) available for conjugation on the target cell binding component and the toxin component.

As stated above, the toxin component can be isolated from natural sources, but it is more preferred that the toxin component be formed by recombinant DNA technology. Recombinant DNA technology is well known to those skilled in the art and comprises the following steps: extraction of mRNA, generation of cDNA libraries, generation of vectors, transformation of host cells, and culturing of transformed host cells during which the toxin polypeptide is expressed. Prokaryotic cells (such as E. coli cells) and eukaryotic cells (such as yeast cells) can be used as host cells to express the desired toxin components. These operations are described in more detail below in the preparation of conjugates.

As stated above, the target cell binding component is usually an antibody or antibody fragment or derivative or contains such material. Preferably, the antibody may be a monoclonal antibody. The antibodies may be prepared by known methods; For example, antibodies can be isolated from serum, spleen and hybridomas that select antibodies.

For example, the preparation of antibodies is described in Methods in Enzymology, Vol. 178; Langone, J. (ed.), Antibodies, Antigens and Molecular Mimicry (Academic Press Inc., Part II, "Engineered Antibodies"); Methods in Enzymology Volume 173; Langone, J., and Van Vunakis, H. (eds.), Volume B, Immunochemical Techniques (Academic Press Inc., Part I, "Production of Antibodies").

If the coupling agent contains a -SX-CO- moiety, the desired derivative of the antibody (or toxin component) can be prepared by reacting the antibody (or toxin component) with a compound of the formula Lj-SX-CO-L ₂ - wherein X is above, and denote L [and L _2] leaving group. The -CO-L ₂ moiety may be, for example, a carboxyl group or, more preferably, a carboxyl group converted to a reactive derivative such as a carboxylic acid ester or a carboxylic acid anhydride. The L ₂ group may be, for example, an imidazolyl group or a succinimidyl ester group. L | the thio group may be oxidized by a leaving group such as pyridyl. The coupling agent is usually reacted with a reduced thio group on the antibody or toxin component (preferably the toxin component).

When other coupling agents are used, they are reacted with the antibody or toxin component in a known manner. The coupling agent is usually reacted with the available functional groups on the target cell binding component and the toxin component. However, the target cell binding component and the toxin component can also be directly linked together by forming a covalent bond, such as a disulfide bridge.

If the toxin component contains a sulfhydryl group (such as the sulfhydryl group of the cysteine residue), this group can be used in many ways to form the conjugate. For example, a disulfide bond may be formed between the sulfhydryl group and the sulfhydryl group (or the sulfhydryl group of the coupling agent) of the target cell-binding component. If the toxin component does not normally contain a sulfhydryl group, the toxin component can be converted to a sulfhydryl-containing derivative to facilitate conjugation. For example, iminothiolane may be used as the reagent to form the derivative.

When used as coupling agents, polypeptides may be reacted with the free carboxylic or amino groups of the toxin component and the target cell-binding component.

When the ricin-At-CO-CH ₂ -CH (CH ₃ ) -S- linker used as the toxin component is to be coupled to the target cell-binding component, for example, N-succinimidyl-3- (2-pyridyldithio) butyrate may be used. Preferably, immunotoxins of this type are prepared by reacting a derivative of the target cell binding component with a coupling agent with reduced ricin A.

The target cell-binding component is reacted with a coupling agent such as N-succinimidyl-3- (2-pyridyldithio) butyrate and ricin A is reacted with the free cysteine 8

The reaction is carried out with a reducing agent (e.g., dithiothreitol) to reduce the thio group on the group.

The coupling agent is usually used in excess; accordingly, a coupling agent derivative of the target cell-binding component has more than one linker group. After the reaction with ricin A, the non-ricin A linker groups are terminated by reacting with a blocking compound such as cysteine.

As stated above, the conjugates of the present invention can be used to treat tumors. The biological activity of the compounds was evaluated in vitro by assaying the inhibitory effect on the protein synthesis of tumor cells and / or by measuring in vivo the effect of reducing the size or growth of the tumor. The course of the biological assays and their results are described below.

The invention is further illustrated by the following non-limiting Examples. In the examples, reference is made to the accompanying drawings, in which Figure 1 illustrates the construction of pICI 0020; Figure 2 illustrates the construction of pTB344; Figure 3 illustrates the construction of pICI 0042; Figure 4 illustrates the construction of pICI 1079; Figure 5 illustrates the construction of pICI 1187; Fig. 6 shows a SDS gel electrophoresis of E. coli lysates stained with Coomassie blue; lane A is the lane of pICI 1102, lane B is the lane of pICI 0020, and lane C is the lane for molecular weight markers;

Figure 7 shows a profile of the gel electrophoretogram of pICI 1102, where the peak R corresponds to ricin A;

Figure 8 is a western spot chromatogram of ricin A produced by pICI 1102; lane 1 is the lane for molecular weight markers, lanes 2 and 3 are the non-ricin producing clones, lane 4 is the pICI 1102 band, and lane 5 is the pICI 0020 (non-ricin-A control plasmid) band;

Figure 9 shows a partial sequence of pICI 1102;

Figure 10 illustrates the construction of pICI 1102; Figure 11 shows the structure of the fragment used in plasmid construction;

Figure 12 is a map of plasmid pICI 0042; Fig. 13 shows the sequence of an tamse transcription terminator;

Figure 14 is a map of plasmid pICI 1079; Figure 15 shows the endocytosis curves of ¹²⁵ I-C242 in Colo 205 cells;

Figure 16 shows the immunotoxin C242 / ricin-A Colo

205 illustrates a cytotoxicity curve for cells;

Figure 17 illustrates an antibody; Figure 18 illustrates the cDNA sequence of the C242 kappa chain variable region (cDNA in pKGE761), and Figure 19 illustrates the cDNA sequence of the C242 heavy chain variable region (cDNA in plasmid pKGE762).

Unless otherwise stated in the examples, (i) the evaporation was carried out on a rotary evaporator under reduced pressure, (ii) the operations were carried out at room temperature (i.e., 18-26 ° C), (iii) the yields given are for information only and not necessarily represent the maximum available;

(iv) NMR spectra were recorded at 200 MHz in DMSO-d ₆ solvent and chemical shifts were reported in ppm units relative to the internal standard tetramethylsilane;

(v) the acronyms have the following meaning:

BSA: bovine serum albumin

PBS: phosphate buffered saline

IPTG: isopropylthio-β-galactoside

SDS-PAGE: Sodium dodecyl sulfate (SDS) poly (acrylamide) gel electrophoresis (PAGE)

Ricin-A / C242 immunotoxin preparation

A solution of 4.4 mg / ml of C242 monoclonal antibody in phosphate buffered saline (150 mmol sodium phosphate, pH 7.2) was concentrated to 12 mg / ml by membrane filtration on Amicon YM10 membrane at 4 ° C. . The concentrate was diluted with 0.5 volumes of borate buffer (100 mM sodium borate, pH 9.1). The protein concentration of the resulting solution was determined by measuring the absorbance at 280 nm and verifying that the pH of the solution was 8.8 ± 0.1.

The coupling agent was prepared from N-succinimidyl-3- (2-pyridyldithio) butyrate in anhydrous, double-distilled dimethylformamide or acetonitrile at a concentration of 10 mg / ml. An aliquot of the solution (0.352 mL) containing N-succinimidyl-3- (2-pyridyldithio) butyrate (3.52 mg) was added immediately to the antibody solution. The resulting solution was mixed and allowed to stand at 15 ° C for 1 hour. The solution was pre-equilibrated with a 2.6 x 58 cm desalting column to remove excess reagents and replace the buffer with 50 mM sodium phosphate, 150 mM sodium chloride, 1 mM EDTA, pH 8.0. G25 Sephadex column (manufactured by Pharmacia) at 2 ml / min. Alternatively, the product may be isolated by flow filtration. The eluate fractions containing the desalted antibody derivative were pooled and the protein content of the solution was determined by measuring the absorbance at 280 nm. The extent of coupling agent incorporation was determined by adding an excess of dithiothreitol to the coupled reagent antibody sample and measuring the amount of thiopyridyl groups released at 343 nm. 4-6 moles of coupling agent were incorporated into 1 mole of antibody.

Recombinant ricin A was reduced with excess dithiothreitol in phosphate buffer solution at pH 8 and concentrated by flow-through filtration. To remove excess reagent, Kon9

EN 215 243 Β concentrate in a buffer containing 50 mM sodium phosphate, 150 mM sodium chloride, 1 mM EDTA, pH 8.0, at a rate of 2 ml / min on a 2.6 * 58 cm G25 Sephadex column (provided by Pharmacia). made).

Recombinant ricin A was conjugated to the coupled reagent antibody by mixing the solution containing the recombinant ricin A / coupling reagent in a 1: 1 ratio by weight with the coupled reagent antibody solution (190190 mg reagent was used). To the mixture was added 20% (v / v) glycerol and the reaction vessel was purged with argon. The resulting solution was heated at 15 ° C for 40-65 hours.

Cysteine was then added to a final concentration of 0.2 mmol (at least a 10-fold molar excess relative to the antibody-attached linker groups) to seal the unreacted linker groups on the antibody. The solution was mixed and kept at 20 ° C for 2-3 hours. The completion of the cysteine reaction was checked by treating a sample of the cysteine-treated immunotoxin with excess dithiothreitol and measuring the amount of thiopyridyl groups released at 343 nm. The reaction is complete when no release of thiopyridyl groups is observed.

The conjugate solution was concentrated by membrane filtration on an Amicon YM10 membrane and the concentrate was pre-equilibrated in a 2.6 * 50 cm chromatography column equilibrated with 50 mM sodium phosphate, 25 mM sodium chloride and 1 mM EDTA. (manufactured by Pharmacia) at 3 ml / min. This procedure separated the immunotoxin and unconjugated antibody from unconjugated ricin A. Fractions containing the mixture of immunotoxin and antibody were pooled and the protein content of the solution was determined by measuring absorbance at 280 nm.

The resulting solution containing the unconjugated antibody and immunotoxin was adjusted to pH 6.3 with 1N aqueous hydrochloric acid and applied to an 8 x 26 cm column containing a triazine dye matrix (Mimetic A6XL from ACL plc (Cambridge, UK). flow rate of 1.5 ml / min. The column was washed with 100 ml of starting buffer to dissolve the unconjugated antibody and then eluted with 0.5 molar sodium chloride in stock buffer. The solution containing the immunotoxin was dialyzed against phosphate buffered saline, filtered through a 0.22 micron pore size filter, and the filtrate was stored at 4 ° C.

The purity of the resulting immunotoxin was tested by SDS polyacrylamide gel electrophoresis. According to the test, 50-60% of the total weight of the 45 mg product is mono-ricin A derivative, 10-30% is di-ricin A derivative, 5-15% is tri-ricin A derivative and 10% less than a fraction of unconverted antibody.

Alternatively, gel filtration (removal of residual recombinant ricin A) and dye affinity chromatography (removal of residual C242 antibody) can be included in the purification step. In this case, the reaction mixture obtained during conjugation is diluted with distilled water until the conductivity is less than 5 mS after blocking the unreacted coupling molecules with cysteine. The resulting solution was adjusted to pH 6.0 with 1 N aqueous acetic acid and the solution clarified by membrane filtration. The solution is then applied to the dye-ligand column (2.5 mL of gel per mg of recombinant ricin A used for conjugation). The column was washed with 0.68% sodium acetate buffer pH 6.0 until complete dissolution of unconjugated C242 antibody. The immunotoxin and the remainder of the recombinant ricin A were then eluted with 20 mM sodium phosphate buffer, pH 7.0, containing 0.5 molar sodium chloride. The resulting eluate was concentrated by flow filtration in a spiral-lined ultrafiltration apparatus and applied to a Sephacryl S300HR column (1 mg, 2.5 mL of gel originally used for conjugation reaction; less than 5% of total volume of concentrate). The column may be equilibrated with any suitable buffer system, such as phosphate buffered saline (pH 7.2).

Preparation of coupling agent

Preparation of N-Succinimidyl (R) -3- (2-pyridyldithio) butyrate

(R) -3- (2-Pyridyldithio) butyric acid (90.0 g, 0.393 mol) was dissolved in dichloromethane (630 mL) at room temperature with stirring. To the solution was added N-hydroxysuccinimide (45.2 g, 0.393 mol, 1.0 molar equivalent); the residue was washed with 90 mL of dichloromethane in the flask. After stirring for 30 minutes at -2 ° C, a solution of dicyclohexylcarbodiimide (81.08 g, 0.393 mol, 1.0 molar equivalent) in dichloromethane (540 mL) was added over 35 minutes while maintaining the temperature at 0 ± 2 ° C. We were holding C. The reagent was washed with 90 mL of dichloromethane into the flask. The solution was stirred at 0-10 ° C for 3.25 hours and then allowed to warm to room temperature. The precipitated solid dicyclohexylurea was filtered off and the filter cake was washed with dichloromethane (2 x 180 mL). The filtrate was concentrated under reduced pressure at 20-22 ° C. The resulting brown oily residue (150 g) was chromatographed on a 230-400 mesh 1180 g silica gel 60 silica gel column (pre-soaked in 80:20 toluene / ethyl acetate). Toluene / ethyl acetate (80:20, v / v) was used as eluent. The eluate was evaporated at 0.25 mbar at 28 ° C. 52 g of N-succinimidyl (R) -3- (2-pyridyldithio) butyrate were obtained as a pale yellow gum.

NRM-NMR (270 MHz, CDC1 _3): 1.5 (d, 3H), 2.85 (m, 4H), 2.8 (dd, 1H), 3.2 (dd, 1H), 3, Δ (m, 1H), 7.1 (m, 1H), 7.7 (m, 2H), 8.5 (d, 1H).

The molecular weight of the product as determined by electron impact mass spectroscopy is 326; the fragments formed corresponded to the structure of N-succinimidyl (R) -3 (2-pyridyldithio) butyrate.

HU 215 243 Β

The starting material was prepared as follows:

(a) Preparation of (R) -3-hydroxybutyric acid:

To 1.670 kg (14.152 moles) of (R) -3-hydroxybutyric acid methyl ester while cooling with ice-ethanol, 4698 ml of water followed by 965 ml (16.98 moles, 1.2 molar equivalents) of 47% w / w aqueous sodium hydroxide solution were added. was added at a rate such that the temperature of the mixture was maintained at 0 ± 2 ° C. The reaction mixture was stirred for an additional 1.5 hours at 0 ° C. Concentrated aqueous hydrochloric acid (1490 mL, 17.28 mol, 1.22 molar equivalent) was added over 35 minutes while maintaining the temperature at 5 ° C. The final pH of the mixture was 1.5. Sodium chloride (1061 g) was then added and the aqueous mixture was extracted with methyl tert-butyl ether (3.5 x 10 liters). The extracts were combined and the solvent was distilled off at room temperature under reduced pressure. To the residue was added toluene (7.5 L) and the volatiles were evaporated under reduced pressure. 1189 g (81%) of (R) -3-hydroxybutyric acid are obtained in the form of an oil. This oily substance was cooled to 4 ° C and seeded with a nucleating agent. A solid was obtained.

b) Preparation of (S) - $ - butyrolactone:

To 530 g (5.096 mol) of (R) -3-hydroxybutyric acid was added 1322.8 g (1488 ml, 8.154 mol, 1.6 molar equivalent) of triethyl orthoacetate at 25-30 ° C. A cloudy solution is formed (the turbidity of the solution is presumably due to sodium chloride not removed in the previous step).

Volatiles were evaporated from the solution at 30 ° C and 10 mm Hg. The resulting liquid (2S, 6R) -2-ethoxy-2,6-dimethyl-1,3-dioxan-4-one (1001 g) was heated at 80 ° C for 2 hours and then cooled to 40 ° C. The crude product was distilled through a 50 cm long, 3.5 cm diameter glass column filled with glass Raschig rings.

57.3 g of (S) -B-butyrolactone are obtained; boiling point: 59-62 ° C / 10-12 mbar.

c) Preparation of (R) -3-mercaptobutyric acid:

To 91.12 g (1.058 mol) of (S) -P-butyrolactone was added 80.57 g (75.65 ml, 1.058 mol, 1.0 molar equivalent) of thioacetic acid under stirring and cooling at 5 ° C under nitrogen. the temperature of the mixture was maintained at 5 ° C. Triethylamine (146.7 mL, 1.058 mol, 1.0 molar equivalent) was then added over 45 minutes while maintaining the temperature between -5 ° C and + 5 ° C. The cooling bath was removed and the mixture was heated to 65 ° C. An orange solution was formed. The reaction mixture was cooled to 10 ° C and stirred at room temperature overnight. The solution was cooled to 0 ° C and water (93 mL) was added. The mixture was further cooled to -10 ° C and a mixture of 185 ml of 47% aqueous sodium hydroxide solution and 92 ml of water was added at such a rate that the temperature was -10 ° C to + 10 ° C. left. The mixture was stirred at -5 ° C for 1.5 hours and then allowed to warm to room temperature. The mixture was washed with methyl tert-butyl ether (265 mL). The aqueous phase was diluted with 185 mL of water, cooled to 0 ° C, and concentrated aqueous hydrochloric acid solution (270 mL of hydrochloric acid solution) was added at 0-10 ° C to pH 1 to 2. The aqueous mixture was then extracted twice with 265 ml of methyl tert-butyl ether. The organic extracts were combined, washed with water (265 mL), dried over anhydrous magnesium sulfate and evaporated under reduced pressure at 30 ° C. The resulting 132.8 g of an orange oily residue was purified by distillation under reduced pressure. 86.6 g of (R) -3-mercapto-butyric acid are obtained; boiling point 96-100 ° C / 2 mbar.

(d) Preparation of pyridine-2-sulfenyl chloride:

2,2'-Dipyridyl disulfide (110.0 g, 0.5 mol) was dissolved in dichloromethane (800 mL) at room temperature and the resulting pale yellow solution was cooled to 0 ° C. Chlorine gas was introduced into the solution at 0-5 ° C. After 3 hours 25 minutes, the solution is saturated with chlorine gas and changes to a golden color. The resulting solution was concentrated under reduced pressure at 0-5 ° C and excess chlorine was removed. 11.8 g of pyridine-2-sulfenyl chloride precipitated as a yellow solid were filtered off and washed with dichloromethane. A further 133 g of pyridine-2-sulfenyl chloride can be isolated from the dichloromethane filtrate (1624 g).

e) Preparation of (R) -3- (2-pyridyldithio) butyric acid:

The yellow solid (11.8 g) and a solution (1035 g, 0.667 mol total pyridine-2-sulfenyl chloride, 1.0 molar equivalent) obtained above were diluted with dichloromethane (50 ml). The resulting mixture was stirred at room temperature for 5-10 minutes, then cooled to -5 ° C and (R) -3-mercaptobutyric acid (80.0 g, 0.667 mol) was added at -5 ° C to 0 ° C. of dichloromethane. The mixture was stirred overnight at room temperature. The mixture was partitioned between brown oil and yellow upper phase. Water (485 mL) was added, cooled to 5 ° C, and the pH of the mixture was adjusted to 4.55 with 385 mL of 2N aqueous sodium hydroxide. The dichloromethane layer was separated and the aqueous layer was extracted with 200 mL dichloromethane. The dichloromethane solutions were combined, dried over anhydrous magnesium sulfate and evaporated under reduced pressure. The resultant pale bamboo oily residue was vigorously mixed with 325 mL of hexane and cooled in an ice bath until a solid formed. After stirring for three hours, the precipitated brown crystalline solid was filtered off, washed with hexane (2 x 162 mL) and dried under reduced pressure at room temperature overnight. 140 g of (R) -3- (2-pyridyldithio) butyric acid are obtained.

NMR (270 MHz, CDC1 _3): 1.45 (d, 3H), 2.6 (dd, 1H), 2.8 (dd, 1H), 3.4 (m, 1H), 7, 1 (m, 1H), 7.7 (m, 2H), 8.5 (m, 1H).

Preparation of C242 (C242: II) antibody

Development of a hybridoma cell line and production of C242: I1 antibody

Production of advanced spleen cells:

Colo 205 human colorectal carcinoma cell line (CCL 222 cell line available from American Type Culture Collection (ATCC), Rockville, Maryland, United States) Iscove's Modified Eagle's Medium supplemented with 10% fetal calf serum

HU 215 243 Β donated. Cells were harvested prior to confluence (usually 2-3 days after sub-culture) and washed three times with phosphate buffered saline for immunization.

Mice of BALB / c strain 4-6 weeks were immunized with an intraperitoneal dose of 3 x 10 ⁷ Colo 205 cells in 0.1 ml phosphate-buffered saline. The animals were subsequently re-injected with 3 x 10 ⁷ Colo 205 cells in suspension of 0.1 ml. After 4 days, the animals were sacrificed, the spleen of the animals was removed and a single cell suspension of the spleen was prepared.

Production of the hybridoma:

The 1.2 * 10 ^s spleen cells from the above cell suspension were split into two test tubes. For each tube, a cell line available from the Sp2 / 0 mouse myeloma cell line (American Type Culture Collection (ATCC) strain collection, Rockville, Maryland, USA); 10 ⁸ myeloma cells from CRL 1581] were added. The test tubes were centrifuged and the fluids were discarded. To each tube was then added slowly 2 ml of a 37 ° C polyethylene glycol solution (10 g of 4000 molar polyethylene glycol), 1 ml of dimethyl solution containing sulfoxide and 10 ml of monoclonal saline]. The tubes were placed in a 37 ° C water bath for 90 seconds while the contents of the tubes were gently stirred. To terminate the fusion, 2 mL of each solution was added to each tube over 30 seconds, 6 mL of physiological buffer was added over another 30 seconds, and an additional 32 mL of physiological buffer was added over 1 minute. The cell suspension was washed with the medium and then suspended in a total of 100 ml of medium supplemented with a mixture of hypoxanthine / aminopterin / thymidine (HAT). The medium used was Iscove's modified Eagle's medium supplemented with 10% fetal calf serum containing 36 mg / L L-asparagine, 116 mg / L Larginine.HCl, 10 mg / L folic acid, 292.3 mg / L L- glutamine, 110.1 mg / L sodium salt of pyruvic acid and 3.49 μΐ / ΐ mercaptoethanol. The 50 μΐ aliquots of the resulting suspension were divided into 96-well tissue culture wells. The wells were layered with 250 μrof BALB / c mouse macrophage suspension (2x1ο ⁴ cells per ml of HAT supplemented medium). The medium was changed 6 days after fusion.

Screening for antibody hybridomas:

Depleted media obtained after 6 days as described above by Kennett R. H. (Kennett, R. H., McKelven T. J., Bechtol K. B. (eds.): Monoclonal Antibodies, Hybridomas; A new dimension in biological assays (Plenum Press, New York, 1980), entitled "Enzyme-linked antibody assay with cells attached to polyvinyl chloride plates", was tested for the presence of antibodies positive for Colo 205. The wells of the Nunc type IIF Nunc immunological plate were coated with 2 x 10 Colo 205 cells per well (50 μΐ 1 mg / 100 ml cell suspension in phosphate buffered saline). Subsequently, 50 μΐ per well of the six-day-old depleted medium was added to the coated wells. Plates were incubated overnight at room temperature and then plated in 100 µl / well of 1% bovine serum albumin diluted in 1: and the plates were incubated overnight at room temperature. Subsequently, 100-100 μΐ of substrate solution (4 Dako S2000 OPD tablets dissolved in a mixture of 12 ml of citric acid buffer pH 5.0) and 5 μΐ of 30% hydrogen peroxide solution were added to the wells and 450 incubated. Absorbance was measured at nm.

Roughly 600 hybridoma clones were analyzed, of which 190 reacted with Colo 250 cells. Of the latter, 177 clones also reacted with normal human lymphocytes (the reagent was prepared from Nyegaard A / S (Norway), Lymphoprep); these clones were omitted from further investigations. The clone developed from one of the remaining clones was called the C242 hybridoma; this clone belongs to the IgG1 class. The C242 hybridoma was subcloned in several steps to form a stable monoclonal antibody producing C242: II.

Multi-step subcloning was performed as follows:

First cloning of the C242 hybridoma resulted in C242: 5 clone. This clone was further cloned to form C242: 5: 2. During both cloning, the hybridoma suspension was diluted to 4 cells / ml in medium supplemented with hypoxanthine / thymidine (HT). 96-well culture plates were coated with a 250 μΐ suspension of BALB / c mouse macrophage in HT supplemented media (5 x 10 ³ macrophages per well) and 50-50 λ of the hybridoma suspension was added to the wells of the plate ( , 2 cells / well). After 2-5 days, single cell clones were detected by microscopic examination. On day 6, the medium was changed and aliquots of depleted medium were assayed for the presence of antibodies that bind to Colo 205 cells in the enzyme-linked assay described above (ELISA) but not to normal human cells. The most favorable clones for the binding strength of Colo 205 cells and the lack of binding to normal human cells were selected and stored for further operations frozen in glass vessels under liquid nitrogen.

For the third cloning, cells were allowed to thaw and cultured in Dulbecco's modified Eagle's medium (containing 5% fetal calf serum). Cells were then plated at an average density of 3 cells / well in wells of a tissue culture plate with 96 wells. Cloning was carried out in the same medium using mouse macrophages as the feed cells. The plate contained 30 clones, 16 of which were nephelometrically tested for IgG production and ELISA for antibody specificity. 12 of the clones produced IgG. The 10 clones thus obtained were further cultured on a tissue culture plate with 24 wells, and the depleted medium was used for IgG production and

HU 215 243 Β specificity. This selection yielded 4 clones with relatively high productivity. The productivity of the four clones obtained was examined by duplicate culture in tissue culture flasks. The highest productivity clone was isolated and stored frozen in liquid nitrogen until further use. This clone is designated C242 / CL 510 Al.

After the first cloning of clone C242 / CL 410 A1, one-third of the cells were found not to produce IgG (absorbance was less than 0.1 at a 1: 1000 dilution in the general mouse IgG ELISA). Clone C242: I was then formed by combining five positive subclones. Productivity of the clone - 150 μg mouse IgG / ml as determined by HPLC.

Clone C242: I was then cloned into 96-well tissue culture plates. 33 clones were obtained, each of which produced murine IgG. Five thus obtained clones having a productivity greater than 150 pg / ml were combined to give the final C242: II clone, which had stable IgG productivity. According to HPLC, the productivity of clone C242: II was 196 pg mouse IgG / ml. The cells were frozen and stored in vials immersed in liquid nitrogen until further use. A sample of the thus produced C242: II hybridoma was deposited under accession number 90012601 in the ECACC cell collection.

Development of C242 monoclonal antibody:

A cell suspension was prepared by thawing a vial containing the above hybridoma. Cells ^χ 2 10 ⁴ C242: 6000 cm ² were seeded total area Nunc A / S type tissue culture spaces II hybridoma cells / ml density. Cells were modified by Iscove [Iscove Ν. N. and Melchers, F., J. Exp. Med., 147, 923 (1978)] and grown on Eagle's medium supplemented with 1% gentamicin, 1% amino acid additive and 5% fetal calf serum. During culturing, the cells were kept at 37 ° C in humidified atmosphere containing 8% carbon dioxide for 4-5 days. After incubation, the cells were counted and samples were taken from the mixture to determine the concentration of monoclonal antibody. The cell culture supernatant was decanted and filtered through a cellulose filter to separate the suspended cells. Subsequently, the supernatant was concentrated to a factor of 20-30 using ultrafiltration in a Millipore Pellican Ultrafiltration cell using a 30,000 exclusion molecular weight membrane. The concentrate was sampled to determine the concentration and antigenic reactivity of C242: II monoclonal antibody and stored at -70 ° C until purification.

Purification of C242 monoclonal antibody:

Protein A-Sepharose 4B gel (manufactured by Pharmacia AB, Uppsala, Sweden) was swollen and washed according to the manufacturer's instructions followed by 1.5 mol glycine sodium hydroxide buffer pH 8.9; buffer solution containing 3 moles of sodium chloride), the gel was bound to C242: II (hereinafter C242 antibody) concentrate prepared as described above. The affinity column was first washed with a buffer solution of the same composition as above, then eluted with 0.1 M citric acid, sodium hydroxide buffer pH 5.0. The eluate fractions containing the C242 monoclonal antibody were collected in test tubes containing 1 M TrisHCl buffer pH 8.0. The resulting antibody solution was dialyzed against 0.02 M phosphate buffer, pH 7.2, containing 0.15 moles of sodium chloride and 0.2 g / l of sodium nitrate. The dialyzed C242 antibody solution was concentrated by ultrafiltration using a 30000 exclusion molecular weight membrane. The concentrate was sampled for protein concentration and quality control analyzes and the C242 monoclonal antibody concentrate was stored at -20 ° C.

PREPARATION OF RICIN-A

Ricin A can be produced by recombinant DNA techniques using DNA sequences encoding ricin A. Such sequences are disclosed in European Patent Application No. 145,111; O. Hare et al., FEBS Letts., 216, 73-78 (1987) and Lord et al., Eur. J. Biochem., 148, 265-270 (1985). The DNA sequence encoding ricin A must be controlled by appropriate control sequences such as a promoter (e.g., a trp promoter), a ribosome binding site, and transcription arresting sequences. The preparation and purification of recombinant ricin A is described in WO 85/03508 and EP 237 676.

The following describes the construction of a recombinant plasmid in which the transcription and translation of the ricin A chain coding sequence is driven by a prokaryotic promoter element. To stop transcription, a natural stop element was inserted at the 3 'end of the information. The initiation of translation is controlled by the DNA sequence at the 5 'end of the information, the ribosome binding site (RBS).

Here's how:

Synthetic oligonucleotides

Synthetic oligonucleotides were used to introduce specific DNA sequence modifications in the ricin gene. All oligonucleotides to be described below were prepared on an Applied Biosystems Type 380A DNA Synthesizer from 5 '- (dimethoxytrityl protected) -nucleoside-2-cyanoethyl-N, N-diisopropylphosphoramidites and 0.2 micronized nucleosides in regulated pore glass supports. . Synthesis was performed according to Applied Biosystems Inc.

The synthesized oligonucleotides were cleaved from the solid support, all protecting groups present were removed, and the resulting oligonucleotides were dissolved in 1 mL of water and the concentration of the solution determined by determining absorbance at 260 nm.

2. Enzymes and host cells

Various restriction endonucleases and DNA modifying enzymes were used in the following operations. These enzymes are from several manufacturers (Amersham International; Bethesda Research

HU 215 243 Β

Laboratories; Boehringer Mannheim; New England Biolabs) and used according to the manufacturer's instructions for reaction conditions.

The host cells used in the operations are generally available from a variety of sources. For example, E. coli strain C600 is freely available from multiple sources, including multiple strain collections; an E. coli C600 is purchased from E. coli GCSC 3004 in the E. coli Genetic Stock Center Collection (Yale University, United States of America). The E. coli C600 genotype is K12 thr-1 leuB6 thi-1 lacYl tonA21 X ~ supE44. E. coli strains HB101 and DH5a are available from Bethesda Research Laboratories, while E. coli strains TG1 and K19 are from the Biotechnology Institute, England.

3. Geneclean ^R Reagent Kit

The Geneclean kit used for the DNA purification procedure described by Vogelstein and Gillespie (1979, 76, 615) consists of the following reagents:

1. 6 M sodium iodide solution; 2. Concentrated solution containing sodium chloride, Tris and EDTA for the preparation of a washing liquid of sodium chloride / ethanol / water; 3. Glassmil ^R : 1.5 ml ampoule containing 1.25 ml of an aqueous silica gel suspension.

However, other methods for purifying DNA may be used, such as those described in Sambrook, Fritsch, and Maniatis, Molecular Cloning - The Laboratory Manual (2nd edition, Cold Spring Harbor Laboratory, 1989). (The latter manual is hereinafter referred to as the Maniatis Manual.)

4. Sequenase ^R

Chemically modified T7 DNA polymerase.

Based on the method of Tabor and Richardson (1987, Proceedings of the National Academy of Sciences USA, 84, 4767-4771).

5. Construction of pIC1 expressing vectors

5.a pICI0020:

Plasmid vector pICI0020 is a pAT153-based plasmid in which the 651 bp EcoRI-AccI region is replaced by a 167 bp EcoRI-ClaI fragment consisting of the following components:

(1) a synthetic E. coli trp promoter and a trp leader ribosome binding site;

(2) a translation initiation codon;

(3) a multiple restriction enzyme recognition sequence comprising Mp3mp18 containing KpnI, BamHI, XbaI, SalI, PstI, SphI and HindIII sites;

(4) a synthetic transcription termination sequence.

The DNA sequence of this region is all. illustrated.

Construction of a plasmid vector containing the synthetic trp promoter sequence is described by Windass et al., Nucl. Acids. 10: 6639-6657 (1982). To isolate the promoter fragment, the vector was digested with EcoRI and Hpal and the appropriate band isolated on the agarose gel was purified by electro-elution (described in the Maniatis manual).

A complementary synthetic oligonucleotide pair was constructed that was ligated to the H-terminus of the promoter fragment and provided the natural trv leader ribosome binding site, the translation initiation codon and the 3 'Kpn1 cloning site. These oligonucleotides were mixed at equimolar concentrations, the chains were assembled by heating to 100 ° C, and the mixture was slowly cooled to room temperature.

The promoter fragment was then ligated to the ligated oligonucleotides and the appropriate band was separated by electroelution from a polyacrylamide gel. The resulting fragment was ligated with the M13mpl8 vector derivative containing the trp damping sequence (a sequence prepared from synthetic oligonucleotides) cloned into the HindIII site and containing an additional Clal restriction site at the 3 'end of the damping sequence. Ligated DNA was CaCl ₂ (Maniatis Manual

Chapter I, p. 82) was transfected into competent E. coli strain JM109 (Yanisch-Perron et al., Gene, 33, 103 (1985)). The transfected microorganism was plated on a culture plate, incubated, and the developed plaques were labeled with ³² P labeled with Benton and Davies (Maniatis Manual, Chapter 4, p. 41) with a labeled translation of a pre-isolated EcoRI-Hpal promoter fragment. screening. Positive hybridizing plaques were generated in a known manner (Maniatis Handbook, Chapter 4, page 29) using single-stranded DNA and the resulting DNA using the M1 universal primer and the Sanger dideoxy chain breaker (a kit containing a number of sources). available from the United States, such as the Sequenase Kit of Bioscience).

RF DNA was prepared from one of the isolates in which the promoter / ribosome binding site / attenuator sequence was identified. This DNA was digested with EcoRI and Clale, and the corresponding fragment was isolated from the polyarylamide gel as described above. Plasmid pAT153 was digested with EcoRI and Steel and ligated with the isolated promoter fragment. The ligated DNA was transformed into competent E. coli HB101 (Boyer H.W. and Roulland-Dussoix D:

J. Mole. Biol., 44, 459 (1969); Bethesda Research Laboratories] and ampicillin-resistant colonies were isolated.

Plasmid DNA was prepared from several different clones and the DNA sequence was analyzed in the region between the EcoRI and ClaI sites. One such clone, which was found to contain the correct promoter / attenuator region, was designated pICI0020.

The construction of the pICI0020 is illustrated in Figure 1.

5.b pICI0042:

Plasmid pICI0042 of the structure shown in Figure 12, in which the antibiotic resistance markers pAT153 is replaced by a single inducible tetracycline resistance gene derived from plasmid RP4 (a substance encoded by the tetA gene and driven by the product of the tetR gene). For the identification of the latter genes Klock et al

HU 215 243 Β, c. Bacteriol., 161: 326-332 (1985)]. Because the novel resistance marker is expressed only in the presence of an antibiotic, the tetA gene product does not appear as a potential contaminant for recombinant ricin A in cultures in which the plasmid is maintained in the absence of tetracycline. The plasmid also contains a plasmid stability function (cer).

In the first step of constructing the vector, a pAT153 derivative was constructed from which the gene encoding tetracycline resistance was completely deleted. To replace the EcoRI-Aval fragment of pAT153, a short sequence complementary synthetic oligonucleotide pair was constructed containing several unique restriction endonuclease sites for further cloning.

Plasmid pAT153 DNA was digested with EcoRI and Aval and the 2.175 Kbp plasmid DNA fragment was isolated from a 0.7% agarose gel with Geneclean (Bio 101, California) according to the manufacturer's protocol. This eliminated the 1.425 Kbp fragment containing the tetracycline resistance gene.

The oligonucleotides SEQ IDs 13 and 14 were phosphorylated with T4 polynucleotide kinase and then equimolarly assembled. The ligated oligonucleotides were then ligated with a plasmid fragment isolated from pATl 53. The ligated DNA was transformed into E. coli HB101 (BRL) and ampicillin resistant colonies were isolated from the culture.

Several colonies were isolated for the construction of small-scale plasmid DNA (Bimbomim and Doly method, described in Maniatis Manual, Chapter 1, page 25), and the desired structure was determined by restriction analysis with appropriate enzymes such as EcoRI, Aval and BamHI. . Three isolates that proved to have the desired restriction structure in this assay were subjected to further structural analysis by DNA sequence analysis using pBR322 primer (left-to-right reactor with EcoRI sites; New England Biolabs). One of the isolates was designated pICIO 19.

Plasmid RP4 DNA was isolated from existing stock using the method of Holmes and Quigley (Maniatis Manual, Chapter 1, page 29). This DNA was cleaved completely with bglll and then with Xmal in detail (reaction at 25 ° C for up to 35 minutes) and samples were taken at various times until the sample was 2.45 Kbp in size. the fragment containing tetR and tetA was not clearly recognizable. The PUC8 DNA sample (Amersham International) was completely digested with Bam HI and Xmal and the tetracycline resistance genes were ligated to pUC9. The ligated DNA was transformed into E. coli C600 cells competent by the CaCl ₂ method (Maniatis Manual, Chapter 1, page 82) (Appleyard RK: Genetics, 39, 440 (1954)) and tetracycline resistant from the strain culture. colonies were isolated. Plasmid DNA was prepared from eight clones by the method of Holmes and Quigley (Maniatis Manual, Chapter 1, page 29) and confirmed by restriction analysis.

RP4 on the presence of tetR and tetA genes. One of the isolates thus obtained was labeled with pTB344.

The tetracycline resistance genes were then inserted into pICI1919 (see above) by replacing an EcoRI / PstI fragment of pICI0019 with the corresponding fragment of pTB344. By this operation, the majority of the ampicillin resistance genes of pIC10019 were replaced with tetracycline resistance genes. Plasmid DNA was transformed into E. coli C600 after digestion and ligation, and colonies were isolated based on the phenotype (Tc ^R and Ap ^s ). Plasmid DNA was prepared from four such clones and digested with enzyme combinations (e.g., BamHI / PstI / SstI, EcoRI / SalI, Smal, Styl / SalI and Aval / PstI). All four clones had a restriction structure corresponding to the desired structure. One of the clones was designated pTB351.

Summers and Sherrat (Cell, 36, 1097-1103 (1984)) have shown that the instability of ColEI-derived plasmids (such as pAT153) is due to the fact that these plasmids do not contain the 283 bp cer sequence present in the original plasmid. . This sequence helps prevent the formation of plasmid oligomers; plasmid oligomers disrupt plasmids through an as yet unclear partitioning mechanism. The cer sequence [Summers et al., Mol. and Gene. Genetics, 201, 334-338 (1985) and Cell, 36, 1097-1103 (1984)] from a cloned fragment of pUC18 (pKS492). Plasmid pKS492 DNA was digested with BamHI and Taq1 to release a 289 bp cer-containing fragment. Plasmid pTB351 DNA isolated from Dam E. coli GM48 host cell (Arraj, J.A. and Marinus, M.G., J. Bact, 153, 562-565 (1983)) was completely digested with BamHI and Clale and then ligated with the digested pKS492 DNA. The ligated DNA was transformed into competent E. coli C600 cells and tetracycline resistant colonies were isolated. The presence of the cer sequence was examined by restriction analysis with clones Aval, Myl and Pvul. One of the isolates which had the desired structure was labeled with pCI0042.

5.c pICI1079:

Plasmid pICI 1079 is an ampicillin-resistant plasmid derived from pAT 153 which contains the following elements between the EcoRI and StyI restriction sites:

(i) the CI857 gene from phage λ;

(ii) the λ P _L promoter;

(iii) a synthetic ribosome binding site;

(iv) synthetic interferon a2 gene sequence;

(ν) T4 tree-derived synthetic transcription termination sequence between the SalI and StyI restriction sites.

The DNA sequence of the transcription arresting sequence is shown in Figure 13 and the structure of pICI1079 is shown in Figure 14.

Plasmid vector pICI 1079 was deposited with the National Collection of Industrial and Marine Bacteria (23 St. Machaer Drive, Aberdeen, Scotland) on February 19, 1991 in accordance with the requirements of the Budapest Treaty.

HU 215 243 Β

Great Britain) in the strain collection under NCIMB 40370.

Plasmid pICI 1079 was used to construct the source of the T4 transcriptional terminator required to construct the pIC1185 clone expressing ricin A (see section 7d). Plasmid pICI1079 was used as starting material to construct pICI1079. PICI 1043 is an analog of plasmid pICI0020 (see section 5a) which contains an expression cassette between the EcoRI and SalI sites containing the APL1 promoter and the interferon a2 gene [Edge et al., Nucl. Acids. Res., 11, 6419-56435 (1983).

A complementary oligonucleotide pair was synthesized to generate a transcriptional terminator from the 32 genes of bacteriophage T4 with cohesive ends 5'SalI and 3'Sphl. This fragment was ligated with a plasmid fragment isolated from pICI 1043 by complete digestion with SalI and SphI. Thus, an intermediate plasmid (pICI1078) was obtained which alternately contained the T4 terminator and the trp damping sequence.

A further pair of complementary oligonucleotides was then synthesized to replace the trp damping sequence in pICI 1078 and the remainder of the tetracycline resistance gene by inserting between the SphI and StyI sites. A unique BamHI site was also inserted into the plasmid vector with the synthetic fragment.

The above procedure is illustrated in Figure 4.

6. Construction of a clone expressing Ricin-A

6. Construction of plasmid pUC8RA DNA

A clone (pUC8RA) containing the cDNA for the production of ricin A was prepared. This clone contains the A chain cDNA from the -74 base number in the leader sequence to the BamHI site in the B chain (base number 857) in pUC8 (Vieria J. and Messing J. Gene 19: 259 (1982)). (Lamb IF, Roberts LM, Lord JM Eur. J. Biochem. 148: 265-270 (1985)). In addition, translational arrest codon was created at the 3 'end of the finished ricin A final codon by site-directed mutagenesis (see O'Hare, M. et al., FEBS Letts., 216, 73-78 (1987)). The entire A chain coding region is contained in a BamHI fragment of the above clone.

Small amounts of plasmid pUC8RA DNA were generated by the above procedure. A diluted sample of the DNA thus obtained was prepared into competent E. coli DH5a cells for subsequent preparation of the stock material [Hanahan, D .: J.Mol. Biol., 166, 557 (1983); cells in Bethesda

Purchased from Research Laboratories and transformed into ampicillin-resistant transformants. Plasmid DNA was prepared from the resulting clone using the modified Bimboim-Doly method (Maniatis Manual, Chapter 1, page 25). DNA samples were digested with BamHI and BanI in separate operations and compared to products obtained after digestion of the original DNA by Lord et al. With the same enzymes after agarose gel electrophoresis. No differences were found between the restriction forms; on this basis, the two DNA samples were considered to be identical.

6.b Subcloning into M13

The product obtained by digestion of pUC8RA plasmid DNA with BamHI was ligated "by shot" with replicative (RF) DNA from phage M13 K9 strain (Biotechnology England). Ligation was performed as described in Maniatis Manual, Chapter 1, page 68. Control ligations were also performed. The ligated DNAs were transformed into the competent TG1 E. coli strain (Gibson, 1984 / Anglian) by CaCl ₂ (Maniatis Manual, Chapter 1, page 82).

Transformation numbers indicated that ligation was effected efficiently. In the offspring, recombinant phages were expected to occur. Recombinant phages are expected to form plaques on plates containing IPTG + X-gal (BRL), which is due to the degradation of the lacZ (β-galactosidase) gene.

In contrast, wild-type phages form blue plaques due to hydrolysis of X-gal by β-galactosidase.

Several plaque plaques were isolated to generate single-stranded DNA. According to the direct gel electrophoresis of the lysed phage suspensions, a phage clone contained a large insert which, upon sequencing, proved to be the coding sequence for the ricin A chain. Although only 182 bases of the complete ricin A coding sequence were identified during sequencing, this fact was considered as sufficient evidence of the presence of the complete ricin A gene. This clone is designated M13K19RA.

6.c Mutagenesis of the M13K19RA clone

In order to construct a KpnI site compatible with pICI expression vectors at the start of ricin-A, SEQ ID No. 15 of SEQ ID No. 15 is highlighted below:

5 '..... ....... GATAACAA0ATATTCC0CAAA 3'

.... ricin leading these- i -— finished ricin-a ---- vencia ....

we need to change as underlined: SEO ID No. 16:

5 »----- GATAACAACATGGTACCCAAA ----- 3 '

J KpnI

Initiate translation

With this modification, a KpnI fragment containing an ATG ricin-A overlapping a KpnI site is generated and this ICI is decoded. The mutant can be inserted into a series of deletion vectors 60 which can be excised. Two N-termi16

HU 215 243 Β (amino acid I-Phe to Met-Val).

For each mutation procedure, single-stranded DNA from M13K19RA was used as a template during mutagenesis. For the mutation operations, a single oligonucleotide DTR16 was synthetically produced, which introduces all mutation changes. The oligonucleotide DTR16 (SEQ ID No. 17) has the structure:

5 * AAGAAGATGGTAGGGAAAGAA 3 '

There are several known methods for introducing specific changes in the DNA sequence by site-directed mutagenesis. In the following operations, Eckstein et al., Nucl. Acid Sl., 13, 8749-8764 (1985) and 14.9679-9698 (1986)], and the kit developed according to that method (manufactured by Amersham International) was used according to the manufacturer's instructions.

The principle of the method is to primer the single-stranded DNA template with the mutagenic oligonucleotide and synthesize a complementary strand containing dATPaS at the site of dATP. Using this nucleotide, phosphorothionate bonds can be formed which some restriction enzymes (such as the Ncil enzyme) cannot cleave. After synthesis of the second strand, the base strand is cleaved with NcII and digested with exonuclease III at the point of mutation. The parent strand is then re-synthesized with DNA polymerase I. Accordingly, the mutagenic oligonucleotide is used as a template for re-synthesis and the mutation is introduced into both strands prior to transformation. The mutation rate is up to 96% in the total progeny population. For screening, plaques are randomly selected for sequence analysis.

In our experiments, all four of the four randomly selected plaques contained the desired mutations.

A mutant (MRA16) was selected to generate RF DNA and assayed for the presence of the newly formed restriction fragment (Kpn1).

Cloning, expression and preliminary identification tests

The pICI expression vectors described in section 5 are capable of introducing DNA fragments cloned into the unique Kpn1 restriction site adjacent to the Trp promoter. The KpnI site overlaps with the translation initiation codon (ATG) 8 bp downstream of the promoter Shine-Dalgarno site (AGGA).

After confirming the structure and sequence of MRA16, RF DNA was digested with Kpn1 on a large scale (about 5 pg RF DNA), and the product was applied to Nu-Sieve GTG agarose gel (manufactured by FMC Bio-products) and manufactured by the manufacturer. following the instructions for use, the DNA fragment encoding ricin-A was isolated by phenolic extraction from the appropriate cut gel slice.

Plasmid PICI0020 (see section 5a) was digested with KpnI followed by calf intestinal alkaline phosphatase (CIP;

Dephosphorylated by Boehringer Mannheim. This treatment prevented the vector from re-circulating during ligation; otherwise, this phenomenon would result in the appearance of a large number of parent individuals in the transformation progeny population.

For ligation, the plasmid vector and the isolated fragment were used in a weight ratio of 8: 1 to 1: 3, depending on the particular procedure. Control ligations were performed to examine the results of phosphatase treatment, ligase activity and the like. The ligation conditions were selected according to the nature of the T4 DNA ligase used (purchased from New England Biolabs or Amersham). The reaction mixtures were usually incubated overnight at 15 ° C.

Half of each ligation reaction mixture (5 μΐ) was diluted to 100 μΐ with 1 xTNE buffer solution (50 mM Tris, 50 mM sodium chloride, 1 mM EDTA) and 200 μΐ of competent E. coli DS410 cells were added. . Transformation was performed as usual (Maniatis Manual, Chapter 1, page 74), and cells were plated on L agar plates supplemented with 25 μg / ml streptomycin and 100 μg / ml ampicillin and incubated overnight at 37 ° C. . The plates were then examined. Plates treated with the ligation reaction usually developed 5-10 times more colonies than non-ligated control plates. In some cases, the number of colonies obtained after culturing in the presence or absence of the ligase differed only slightly, indicating that the digestion of the vector was incomplete or that the activity of the ligase was low.

For hybridization screening, samples of transformants and appropriate controls were plated on nitrocellulose filters on L agar plates (performed by Grunstein and Hogness; see Maniatis Manual, Chapter 1, page 98). After incubation, colonies were dissolved in situ using 10% SDS and 1 molar sodium hydroxide solution, neutralized with 1 molar Tris buffer, pH 7.5, and dried under reduced pressure at 80 ° C for 2 hours.

Mutation oligonucleotides were labeled with ³² P using T4 polynucleotide kinase to generate hybridization probes. The filters were treated with the probes at room temperature and the non-specifically bound radioactive material was washed in several steps at temperatures ranging from 55 to 65 ° C. The filters were then subjected to autoradiography. Clones detected by specific hybridization contain ricin-Α DNA.

Small-scale DNA preparations were made from positively hybridizing clones using the Holmes and Quigley method or the Bimboim-Doly method (Maniatis Manual, Chapter 1, page 25). The DNAs were digested with the appropriate restriction enzymes, i.e. KpnI and EcoRI / BglII, and the products were analyzed by agarose gel electrophoresis. The vector DNAs and mutation modified RF DNAs were cleaved with the same enzymes to determine the expected fragment sizes in the correct structure clones.

HU 215 243 Β

Large-scale plasmid DNA preparations were prepared from each clone by the method of Bimbóim and Doly for more detailed restriction analysis. For example, restriction enzymes Clal, HindIII, BamHI, EcoRI / BglII, KpnI and Seal were used for restriction analysis. The resulting products were analyzed on agarose gels. The analysis results indicate that the size, location and restriction enzyme cleavage sites formed are unique to the ricin A chain gene.

6.e Expression Tests

Clones found to be positive by hybridization assay and restriction screening were analyzed by SDS-PAGE of whole cell lysates for expression of ricin-A. Expression assays were performed under the following conditions:

1. 10 ml of L-medium supplemented with antibiotic / antibiotics was inoculated with a single colony and grown at 37 ° C with gentle shaking overnight.

2. An aliquot of 750 μl of L-medium was separated and centrifuged at 6500 rpm for 1 minute in a microcentrifuge.

3. The cell pellet obtained by centrifugation was resuspended in 300 μl M9 medium supplemented with 0.02% casein hydrolyzate, 0.2% glucose, and 50 pg / ml thiamine (Maniatic Manual Appendix A.3), and 10 ml of suspension was added to the suspension. medium with the same composition was inoculated.

4. The culture was incubated for 7 hours or overnight at 37 ° C with gentle shaking.

5. After incubation, the OD _{540 of} the culture was measured, the cells pelleted by centrifugation, and the pellet was resuspended in a volume of Laemmli sample buffer (Maniatis Manual, Chapter 18, page 53) as the OD _{54 in} water. would have a value of 10. The resulting suspension was refluxed for 15 minutes.

6. 20 μl of the lysate thus obtained was applied to SDS polyacrylamide gel, electrophoresed, stained with Coomassie blue, and the dye removed and the bands visible.

Only one of the clones examined by SDS-PAGE showed an additional band indicating the presence of a substance having a molecular weight of about 29 kilodaltons (equivalent to the total molecular weight of unglycosylated ricin A). According to gel electrophoresis, this fraction represented 5-10% of the total cellular protein. The plasmid in the particular clone was designated pICII102.

The construction of plasmid pICII102 is shown in Figure 5 and the results of the expression studies are shown in Figures 6 and 7.

6.f. Western blot analysis and immunological detection of recombinant ricin A

The authenticity of the recombinant ricin A protein detected by SDS-polyacrylamide gel staining with Coomassie blue was confirmed by Western blot analysis. The protein bands were transferred to nitrocellulose filters and assayed with ricin A-specific antibody followed by peroxidase-labeled antiglobulins.

15% SDS-PAGE was performed overnight at 8 mA and the gels were allowed to equilibrate with the transfer buffer for at least 30 minutes.

The protein bands on the gels were then transferred to Hybond-C nitrocellulose membranes (manufactured by Amersham). Transmission was performed electrophoretically by treating the bands in a Bio-Rad Trans Biot instrument at 70 V for 3 hours. After drying, the membrane filters were stored in sealed plastic bags at -20 ° C.

In rabbits, a ricin A-1 antibody (polyclonal antibody) was developed against a synthetic peptide fragment of ricin A. Preliminary studies have shown that this antibody has good affinity for ricin A, but that it crosses significantly with many E. coli proteins. The antibody was preincubated with E. coli lysate to eliminate the interference caused by the cross-reactions.

E. coli DS410 was grown overnight in L-medium and 10 ml of culture was centrifuged at 4000 rpm for 10 min. The resulting cell pellet was resuspended in 5 mL of bacterial buffer and irradiated with 4-6 μί wavelengths six times for ten seconds (30 seconds of ice cooling between each irradiation).

0.5 ml of the resulting mixture was mixed with 0.5 ml of ricin A-1 antiserum and incubated at room temperature for 90 minutes. Cell debris was removed by centrifugation at 13,000 rpm for 5 minutes and the supernatant was stored at -20 ° C.

Nitrocellulose membrane filters were incubated overnight in PBS / Tween 5% bovine serum albumin (phosphate buffered saline containing 5 mL Tween 20) for blocking with the proteins applied thereto and washed three times for 3 minutes with PBS / Tween.

Membrane filters were incubated with "blocked" ricin A-1 diluted 1: 4000 in PBS / Tween 0.5% bovine serum albumin for 2 hours (or overnight) at room temperature and washed three times for 3 minutes with PBS / Tween.

Membrane filters were incubated with goat anti-rabbit antiserum diluted 1: 1000 in PBS / Tween 0.5% bovine serum albumin for 1 hour at room temperature and washed three times for 3 minutes with PBS / Tween.

Membrane filters were incubated with rabbit peroxidase anti-peroxidase antiserum diluted 1: 5000 in PBS / Tween 0.5% bovine serum albumin for 1 hour at room temperature and washed three times for 3 minutes with PBS / Tween.

The membrane filters were immersed in a solution of 4 mg of 4-chloronaphtol (60 mg) in phosphate buffered saline (20 ml), diluted with hydrogen peroxide (12 μl), diluted to 120 ml with saline. As soon as the bands became visible, the membrane filters were removed, dried and photographed.

HU 215 243 Β

A typical western spot chromatogram is shown in Figure 8.

6.g Investigation of the biological effect of recombinant ricin A

We aimed to establish experimental conditions under which the biological activity of the samples can be assayed for cell-free protein synthesis in vitro.

Allen and Schweet [J. Biol. Chem., 237, 760-767 (1962)] prepared rabbit reticulocyte lysates. Under the experimental conditions, inhibition of protein synthesis in the cell-free system was assayed by determining whether the radiolabeled leucine enters the newly synthesized protein.

6.g.i Test procedure:

The following solutions or mixtures were used for the tests:

Amino acid mixture (stock solution): A solution containing all amino acids except leucine at a concentration of 1 mmol (the pH of the solution was adjusted to 7.4 with sodium hydroxide and the solution was stored at -70 ° C).

Solution A: Solution containing 40 mmol of magnesium acetate, 2 mol of ammonium acetate and 0.2 mol of Tris (the pH of the solution was adjusted to 7.4 with hydrochloric acid and the solution was stored at 4 ° C) .

Solution B: Solution containing 246 mg / ml adenosine triphosphate (ATP; Sigma A5394) and 24.4 mg / ml guanosine triphosphate (GTP; Sigma G8752).

Analytical mixture:

1 ml amino acid mixture 1 ml Solution "Α" 0.1 ml Solution B 103 mg Creatine phosphate 1 mg CK 510 μΐ water 600 μΐ (60 μϋί) L- ¹⁴ C-leucine (New England Nuclear, NEC-279E) Reaction Mixture: 25 μί sample to be examined 12.5 μΐ analytical mixture 25 μΐ rabbit reticulocyte lysate

Blood solution: phosphate buffered saline containing 2 mg / ml bovine serum albumin (BSA)

All studies were performed in duplicate.

12.5 μΐ of the assay mixture was filled into sterile glass tubes as tubes. The first four tubes were filled with 25 μΐ of blank solution per tube and the remaining four tubes with 25 μ μ of test sample. To the first two tubes was added 1 mL of 0.1 M potassium hydroxide solution (blank background). The tubes were immersed in a 28 ° C water bath and allowed to equilibrate.

Rabbit reticulocyte lysate was allowed to thaw from liquid nitrogen temperature, and 25-25 μί lysate was added at 20-second intervals to each tube. When the first tube was incubated for 12 minutes, 1 mL of 0.1 M potassium hydroxide solution was added again at 20 second intervals (thus incubating the contents of each tube for 12 minutes). Then 2 drops per tube

A 20% solution of hydrogen peroxide was added followed by 1 mL of 20% trichloroacetic acid solution per tube. The contents of the tubes were mixed and allowed to stand at 4 ° C for at least 1 hour (or overnight). The precipitate was filtered through 2.5 cm diameter glass fiber discs, washed three times with 4 ml of 5% trichloroacetic acid, and the precipitate was filled into a scintillation sample holder and 10 ml of scintillation fluid (Ready Solv. MP, Beckman) was added. , and after 1 hour, the contents of the sample holder were shaken and the radioactivity of the sample was measured by scintillation.

ó.g.ii Test procedure using E. coli lysates:

The cultures were grown in 10 ml of L-medium overnight at 37 ° C. Aliquots of the cultures were centrifuged in 400 μl volumes at 13,000 rpm for 30 seconds, the resulting pellet was separated, and the supernatant mass was discarded.

The resulting pellets were rapidly frozen in two cycles of solid carbon dioxide / ethanol and allowed to thaw at 37 ° C. The pellets were then treated with 12 μl of 25% sucrose solution in 50 mM Tris-HCl buffer pH 8.0, followed by 4 μl of 10 mg / ml lysozyme solution.

The resulting mixtures were incubated for 15 min under ice-cooling, then 8 µL of 0.25 molar EDTA solution was added and incubation continued for 15 min. Cell destruction was performed osmoticly by diluting the samples with water to a final volume of 400 μί. This procedure yielded mixtures containing 80-100 viable cells per milliliter.

When a 25 μl aliquot of the resulting lysate is added to the assay mixture described above, the level of incorporation of ¹⁴ C-leucine into the newly synthesized protein is approximately 10% of that observed in the blank without the lysate. This result is similar to that achieved with 8 ng / ml ricin A. Dilutions were then made from the E. coli lysate and the assay was repeated. The results of the tests showed that the lysate had to be diluted at least 16 times in order to reduce its effect to the level of the blank.

In order to ensure, as far as possible, that E. coli lysis and E. coli lysates do not reduce the toxicity of ricin A, two control studies were performed. In the first assay, ricin-A of plant origin was added to the 16-fold diluted E. coli cell pellet at a concentration such that the final concentration of ricin-A after cell disruption was 8 ng / ml. The results of the two control studies showed that neither the lysate nor the cell destruction process adversely affects the ricin-A inhibitor activity.

This method was used to verify that biologically active recombinant ricin A was formed from pICII102 and the clones described below.

6.h DNA Sequence Analysis:

PICII102 was analyzed by plasmid DNA sequencing. The analysis was performed according to the method of Zagursky et al., Gene Analysis Techniques Vol. 2, Number 5, but the procedure was modified. The double19

243 243 HU of the plasmid DNA was basically denatured prior to the primer fitting and known sequencing (for example, using the Sequenase kit of the United States Bioscience). The use of an oligonucleotide cleaving the β-lactamase at the 3'-end and several primers in the A chain made it possible to sequence both strands of the promoter and the ricin A gene.

Initial sequencing data confirmed the unexpected result that there was a KpnI fragment between the promoter and the 5 ricin-A coding sequences (SEQ ID No. 20):

KpnI 'AAAAAGGGTATCGACATGGTACCCGGGGATCCACCTGAGGGTGG KpnI

TCTTTGACATTAGAGGATAACAACATGGTACGGAAACAATAG 3 '

This additional KpnI fragment is derived from M13K19RA and contains restriction enzyme cleavage sites as well as a ricin leader portion cloned from pUC8RA. The 5 'region of the ricin A chain contains base changes introduced during mutagenesis. 20

Studying the sequence shows that the first translation initiation codon (ATG) is outside the scope of the ricin-A coding region.

Before the ricin-A initiating codon, there is an in-frame stop codon (TAG) and a putative Shine-Dalgamos sequence (AGGA) that can restart translation from the second ATG.

Subsequent studies confirmed the surprising result that this additional DNA fragment, when compared to the clones from which it was excised, has a beneficial effect on the accumulation of the ricin A chain in E. coli.

The complete DNA sequence of the ricin A gene in pICII102 is shown in Figure 9.

7. Formation of additional ricin-A-expressing clans

Ί. & PICII 102 Mutation of ricin-A clone for subcloning:

From the unexpectedly formed pIC111 120, it is difficult to subclone the two KpnI fragments in the correct arrangement for expression of ricin A. Therefore, according to our experimental design, we wanted to change the internal KpnI recognition site by a base exchange (A to T) 25. This change prevents KpnI cleavage at the site and allows for the subcloning of a single KpnI fragment into a series of pICI expression vectors. By replacing the adenine at the KpnI recognition site with thymine (i.e. 30 GGT7CC at GGTriCC), the first residue of ricin-A remained unchanged (GTA / GTT = Val). We've made the following conversion:

Starting position:

KpnI 53 bp fragment KpnI Bicin-A sequence KpnI! I! _ - - - - - - - - - _ _ _ _ _ _

GGTAGC ATGGTACC IGA GGTACG

Modified Status:

KpnI 53 bp fragment Bicin-A sequence KpnI! ! !

_ __ - - _______, GGTACC ATGGTTCC TGA GGTACC j

KpnI does not recognize it

The oligonucleotide synthesized to induce the above change has the sequence (SEQ ID No. 19):

5'ATAACAACATGGTICCCAAAC A A T A C 3Ά mutation change is underlined.

We planned to clone the mutagenic ricin A fragment into a number of trp expression vectors for comparative expression studies. After cloning into pICI0020, comparing with pICI 1102, we can determine if and how these single effects affect expression.

7.b Mutagenesis:

The MRA16 template was used as a template for mutagenesis; this is an M13 clone containing two KpnI fragments in pICII 102. Following mutagenesis, isolates in which the desired mutation occurred were identified by random sampling and determination of the DNA sequence of the specific binding region of the mutagenic oligonucleotide.

One of the templates produced by the mutation was designated MRA22. DNA sequencing of this template with the complete ricin A coding sequence20

EN 215 243 ára to confirm that the sequence does not contain any other non-specific mutation.

7.c Subcloning:

Mutation modified single-stranded DNAs were transformed into competent E. coli TG1 cells to form unique plaques. Individual plaques were isolated and the replicative (RF) double-stranded DNA was purified by banding with decreasing density cesium chloride / ethidium bromide mixtures. The purified RF DNA was completely digested with KpnI. Cloning was performed by either "firing" ligation of the digested RF DNA to the appropriate KpnI-cleaved and phosphatase-treated expression vector, or specific ligation of ricin A-phagagene pre-purified on an agarose gel. The ligated DNA was transformed into E. coli TG1 or HB101 cells.

Clones containing ricin-A were identified by hybridization screening (using a ³² Pral labeled ricin-A probe obtained by random priming of a KpnI fragment isolated from another ricin-A-containing clone (pIC111121)). Positively hybridizing colonies were further screened by restriction analysis of plasmid DNA; DNA was digested with KpnI once and twice with EcoRI / BglII for restriction analysis. The size of the inserted fragment was determined by digestion with KpnI and the position of the fragment by digestion with EcoRI / BglII.

Clones which contained the ricin A fragment in the correct position for expression in the above assay were further analyzed by SDS-poly (acrylamide) gel electrophoresis followed by Coomassie staining and Western blot analysis. The level of ricin A accumulation in these clones was equivalent to that observed with pICl102.

One plasmid thus isolated was killed with pICI 1131.

7.d Applying Another Transcription Stopping Element:

The transcription arresting elements form a secondary structure at the 3 'end of the mRNAs, which plays a role in the termination of RNA polymerase activity. Secondary structure stability can improve protein accumulation by protecting mRNA at the 3 'end against attack by exonucleases. The secondary structure of the T ₄ transcriptional terminator is considered to be more stable than the secondary structure of all trp suppressors present in the previous ricin A structure.

In these experiments, the trp promoter and the ricin A fragment were excised from pIC111 by digestion with EcoRI and SalI (the latter enzyme cleaving between the 3 'end of the ricin A coding sequence and the trpA transcription terminator). The resulting fragment was isolated by excision from the agarose gel (2% NuSieve GTG Agarose Gel, manufactured by FMC Bioproducts) and purified by phenol and chloroform extraction followed by ethanol precipitation. The purified fragment was ligated with pCI1079 cleaved with EcoRI and SalI. The latter plasmid contains the _T4 terminator between unique Sail and SphI sites (see Figure 5c).

The ligated DNA was transformed into competent E. coli HB101 cells and assayed for the presence of ricin A DNA by hybridization screening as described above. The positively hybridizing clones were subjected to restriction analysis using EcoRI and SalI together to show that they contained the appropriate size fragment. Positively hybridizing clones of appropriate structure were used to generate plasmid DNA.

One of the clones thus isolated, having the correct structure, was designated pICI 1185.

7.e Use of other plasmid backgrounds:

To construct another construct, plasmid plCI1185 was subcloned into its expression cassette into pICI0042. Plasmid DNA was prepared from pICI 185 digested with EcoRI and SphI enzymes together and thus excised from the trp promoter (RBSl) ricin-A (MRA22) fragment / T ₄ terminator expression cassette containing. This fragment was isolated as described in 4d and ligated with pICI0042 cleaved with EcoRI and SphI.

The ligated DNA was transformed into E. coli HB101 cells and incubated overnight at 37 ° C. HB101 transformants were plated on tetracycline-containing L-agar plates and colonies were subjected to hybridization screening using ³² P-labeled ricin-A DNA probe.

In both cases, the correctness of the plasmid DNA structure in the positively hybridizing colonies was determined by restriction analysis with EcoRI / Sphl and EcoRI / BglII. The three isolates thus isolated were designated pICII 187.1, pICII 187.2 and pICIII 187.3.

The construction of pICI18 185 and pICI1187 is illustrated in Figure 10.

7.f Select Clans:

Isolated plasmids were transformed into E. coli 71.18 cells [Gronenborn B: Moles. Gene. Gene, 148, 243-250 (1976)] and single colonies were selected for clone selection assays. The resulting whole cell lysates were subjected to electrophoresis on two SDS-polyacrylamide gels; then one of the plates was stained with Coomassie-blue while the other was used for western blot analysis.

The stained gel provided little data on the expression of ricin A, due to the presence of a co-migrating protein from E. coli 71.18. Western blot analysis, however, strongly demonstrated ricin-A expression when compared to both positive and negative control samples.

An isolate thus obtained was fermented to size increase.

Fermentation of recombinant ricin A (a) E. coli strain DS410 (also known as MSD68) was transformed with plasmid pICI1187, the resulting recombinant (MSD1051) was purified and stored in glycerol stock at -80 ° C.

An aliquot of the culture was spread on agar plates containing L-tetracycline overnight at 37 ° C.

HU 215 243 Β and individual colonies were isolated from the culture. A single colony of MSD 1051 was harvested, suspended in 10 ml of L-tetracycline medium, but inoculated with 100 μl aliquots of ten 250 ml Erlenmeyer flasks containing 75 ml of L-tetracycline medium each. The cultures were grown on a rotary shaker for 16 hours at 37 ° C, then the contents of the flasks were pooled and the resulting culture was inoculated with 20 liters of modified LCM50 growth medium with the following composition:

Composition of growth medium (g / l distilled water):

KH ₂ PO ₄ 3.0 Well ₂ HPO ₄ 6.0 NaCl 0.5 Casein Hydrolyzate (Oxoid L41) 2.0 (NH ₄ ) ₂ SO ₄ 10.00 Yeast Extract (Difco) 20.00 Glycerol 35.00 MgSO ₄ -7H ₂ O 0.5 CaCl ₂ -2H ₂ O 0.03 thiamine 0,008 FeSO ₄ / citric acid 0.04 / 0.02 Trace element solution (for composition see below) 0.5 ml

Composition of trace element solution (mg / 10 deionized water):

A1C1 ₃ -6H ₂ O 2.0 CoC1 ₂ -6H ₂ O 0.8 KCr (SO ₄ ) ₂ 12H ₂ O 0.2 CuCl ₂ -2H ₂ O 0.2 h ₃ bo ₃ 0.1 WHO 2.0 MnSO ₄ H ₂ O 2.0 NiSO ₄ -6H ₂ O 0.09 Na ₂ MoO ₄ -2H ₂ O 0.4 ZnSO ₄ -7H ₂ O 0.4

The fermentation was carried out at 37 ° C and the pH of the fermentor was maintained at 6.7 by automatic addition of 6M aqueous sodium hydroxide solution. The dissolved oxygen tension (dOT) was set to 50% air saturation and kept at the set value by automatically changing the agitation speed. The air inlet speed was initially set to 20 liters / minute (1 vol / vol / min) and then increased to 45 liters / min when the agitator speed reached a maximum of about 80-90%.

From time to time, the mixture was sampled from the mixture during fermentation and the optical density (OD ₅₅₀ ) of the sample, the dry weight of the cells and the intracellular accumulation of ricin A were measured. The accumulation of ricin A was measured by electrophoresis of the milky cell lysate of bacteria in the sample on an SDE-poly (acrylamide) gel and then stained with Coomassie blue dye. This definition is well known to those skilled in the art.

After 4.5 hours from the inoculation, yeast extract solution (Difco) at a concentration of 1.7 g / l / h was pumped into the fermentor at a rate of 1.7 g / l / h.

After an hour when the OD ₅₅₀ increased to about 50, but before the fermentation became oxygen-restricted, the bacteria were separated by centrifugation for 30 minutes at 7000 g at 4 ° C in a Sorval RC3B centrifuge and recovered from the bacteria. accumulated protein.

(b) E. coli strain DS410 (also known as MSD68 strain) was transformed with plasmid pICI1187, the resulting recombinant (MSD 1051) was purified and stored as a glycerol stock at -80 ° C.

An aliquot of 100 μΐ was separated from the culture and inoculated immediately with a 600 ml volume of Ltetracycline medium in a 2 liter Erlenmeyer flask. The culture was grown on a rotary shaker for 16 hours at 37 ° C. The contents of the flask were inoculated with a growth medium filled with the following composition:

Composition of growth medium (g / l distilled water):

KH ₂ PO ₄ 3.0 Well ₂ HPO ₄ 6.0 NaCl 0.5 Casein Hydrolyzate (Oxoid L41) 2.0 (NH ₄ ) ₂ SO ₄ 10.0 Yeast Extract (Difco) 20.0 Glycerol 35.0 MgSO ₄ -7H ₂ O 0.5 CaCl ₂ -2H ₂ O 0.03 thiamine 0,008 F eSO ₄ / citric acid 0.04 / 0.02 Trace element solution (for composition see above) 0.5 ml / l tetracycline 10 mg / L

The fermentation was carried out at 37 ° C. The pH of the mixture was adjusted by automatic addition of 2M aqueous sulfuric acid solution and 6M aqueous sodium hydroxide solution so that the pH of the mixture was 6.7 and 6.0 for up to 10 hours after inoculation.

The dissolved oxygen tension (dOT) was set to 50% air saturation and kept at the set value by automatically changing the agitation speed. The air inlet speed was initially set to 20 liters per minute (1 vol / vol / min) and then increased to 45 liters / min when the agitator speed reached its maximum (1000 rpm). When, at later stages of the fermentation, the stirring speed was automatically reduced to about 500 rpm, the air inlet rate was manually reduced to 20 liters / minute. The fermentation was carried out for 23 hours. During this time, the mixture was periodically sampled and examined for optical density (OD ₅₅₀ ), dry cell weight, and ricin A accumulation and distribution in bacterial cells. The accumulation of ricin A was measured by electrophoresis on a total cell lysate of bacteria in the sample, and stained with Coomassie blue dye on an SDS-polyacrylamide gel. This definition is well known to those skilled in the art. The distribution of ricin A in the cell cytoplasm (soluble) and in the inclusion bodies (insoluble) was determined by disrupting the bacterial cells in the sample by ultrasound in a known manner.

HU 215 243 Β

After 4.5 hours from the inoculation, yeast extract solution (Difco) at a concentration of 1.7 g / l / h was pumped into the fermentor at a rate of 1.7 g / l / h.

When the carbon source content of the fermentation broth was exhausted (indicated by a sudden increase in dissolved oxygen tension), a nutrient supplement containing 714 g / l glycerol and 143 g / l ammonium sulfate was added to the fermentor at a rate corresponding to the maximum carbon source requirement of the bacteria. The feed rate of carbon source and ammonium sulfate was maintained unchanged for the rest of the fermentation.

After 1 hour of fermentation, the fermentation broth contained approximately 1500 mg / L soluble ricin A.

Note that E. coli strain DS410 (also known as MSD68 strain) is well known (Dougan and Sherrat, Molecular and General Genetics, 151, 151-160 (1977)). The strain is freely accessible to the public. The strain was deposited on June 7, 1985, in accordance with the provisions of the Budapest Treaty, in the National Collection of Industrian and Marine Bacteria Ltd. (Aberdeen, United Kingdom) under the number 12100.

PURIFICATION OF RECOMBINANT RICIN-A FROM E. COLE

Under optimal fermentation conditions, recombinant ricin-A accumulates as a cell-soluble protein. This protein was isolated by homogenizing the cells with the appropriate buffer solution for the stability of the recombinant ricin A during cell disruption. To ensure the solution stability of the product, the operation was performed on live cells immediately after cell isolation. The solid (cell debris) was removed by centrifugation. For scalability, cell debris was flocculated with a polytenimine imine flocculant, which also precipitates the bulk of nucleic acids in the extract. The supernatant obtained by centrifugation was sterile filtered, concentrated by flow filtration and the protein precipitated with ammonium sulfate. The precipitate from the precipitation of ammonium sulfate was frozen at -70 ° C.

The isoelectric point of recombinant ricin-A is pH 7.3, which is significantly higher than that of many other E. coli proteins. Therefore, the product can be easily purified by ion exchange chromatography. All isolation and chromatographic operations were carried out under favorable conditions for the stability of recombinant ricin A, i.e. below 15 ° C in the presence of dithiothreitol and EDTA. Using dithiothreitol, it was ensured that the free thiol groups remained in reduced state while EDTA was added to suppress air oxidation and proteolysis.

Recombinant Castor — Isolation of A

The cells were removed from the fermentation broth by means of a disc separator equipped with a drain. 25-25 liters of fermentation broth from two fermentors were combined and filled in a 50 liter roller container for a series of storage tanks connected to a separator and a high pressure homogenizer.

The scroll tank was fitted to the system and the fermentation broth was pumped through the centrifuge separator at a rate of 40 liters / hour. The evacuation rate was adjusted to be transparent through the entire separation line when visually inspected through the isolated supernatant spectacle lens. The separator solids (i.e., cells) were treated with 5N aqueous sodium hydroxide solution containing 40 L of buffer solution A (50 mmol sodium dihydrogenorthophosphate, 25 mmol ethylenediaminetetraacetic acid, 5 mmol benzamidine and 2 mmol dithiothreitol). PH 6.3 solution) was resuspended and the suspension was cooled to 8 ° C. The suspended cells were then returned to the scroll container through a homogenizer set at 600 bar operating pressure. The resulting 60 L volume of homogenate was cooled to below 20 ° C and the volume of homogenate was adjusted to 0.5% by addition of 2.5 L of 10% polytenimine solution. The resulting suspension was allowed to flocculate for 10 minutes and then transferred to the storage tank via a centrifugal separator. The clear supernatant was then separated and purified and sterilized by filtration through a thick filter and a positively charged, 0.2 μm membrane filter.

Ammonium sulfate precipitation

The sterile purified supernatant was concentrated to a final volume of 12 liters in a spiral-bed cross-flow filtration apparatus and 2.9 kg of solid crystalline ammonium sulfate was added. The solution was saturated with 40% ammonium sulfate. The solution was allowed to flocculate with gentle stirring at 15 ° C overnight and centrifuged. The resulting slurry was stored at -70 ° C until use.

Re-solubilization and desalting

The precipitate obtained by precipitation with ammonium sulfate was adjusted to pH 6.3 with 14 liters of buffer solution B (50 mM sodium dihydrogen orthophosphate, 25 mM ethylenediaminetetraacetic acid and 2 mM dithiothreite). buffer solution). After 30 minutes, the suspension was clarified by centrifugation and desalted by diafiltration with 70 L of buffer solution B. The conductivity of the solution was measured; the conductivity was reduced to below 3 mS / cm. The resulting desalted solution was clarified by repeated centrifugation and immediately processed.

Anion exchange chromatography

The desalted solution was added slowly into a batch chromatography tank filled with 2 kg of DEAE cellulose. (DEAE cellulose was pre-equilibrated with 60 liters of buffer solution B). was pumped through a column of 11.3 cm diameter and 10 cm height filled with DEAE cellulose equilibrated with buffer solution. The bulk of recombinant ricin A was not bound. The solution containing the recombinant ricin A was collected in a stainless steel container.

HU 215 243 Β

Cation Exchange Chromatography The pH of the recombinant ricin A solution was adjusted to 5.5 with 1 molar orthophosphoric acid solution and treated with 10 L of buffer solution C (25 mmol sodium dihydrogenorthophosphate, 5 mmol ethylenediaminetetraacetic acid and 2 mmol dithio). triturate, adjusted to pH 5.5 with 5 N aqueous sodium hydroxide solution), loaded with equilibrated carboxymethyl agarose, 10 cm in diameter and 10 cm in height. The column binds recombinant ricin A. The column was washed with 10 L of buffer solution C followed by 5 N aqueous sodium buffer containing 25 mmol of sodium dihydrogen orthophosphate, 5 mmol of ethylenediaminetetraacetic acid, 2 mmol of dithiothreite and 100 mmol of sodium chloride. buffer solution adjusted to pH 5.5). The eluate fractions containing pure recombinant ricin A (single peak) were pooled and stored as a sterile solution frozen at -70 ° C until further use. Under these conditions, recombinant ricin A is stable for 1 year.

In the operations described above, the following equipment was used:

Anion exchanger: DE-52, DEAE cellulose (Whatman

Biochemicals)

Cation Exchanger: CM / Sepharose (Pharmacia)

Centrifuge Westphalia CSA-1 Disc Centrifuge (Westphalia)

Homogenizer: APV-Schroeder Leg 60/60 Homogenizer (APV)

Filters: 0057P Thick Filter; ABI

NFZP-Posidyne Membrane Filter (Pali)

Batch chromatography tank:

701 Pharmacia Batch Chromatography Tank (Pharmacia)

DE column: Bioprocess 113 (Pharmacia)

DM Column: K100 / 50 (Pharmacia)

INVESTIGATION OF THE BIOLOGICAL PROPERTIES OF THE C242 ANTIBODY AND IMMUNOTOXIN

Immunohistochemistry of C242 (C242.II) in Colorectal Cancer and Normal Tissue

Patients awaiting surgical resection were isolated from primary colorectal cancer tissues and various normal tissue samples. Tissue samples were kept on ice and frozen in liquid nitrogen pre-cooled with isopentane for 1 hour after resection. Tissue samples were stored at -70 ° C until cut. Frozen tissue samples were cut into 5 µm slices and treated with 50% acetone at 4 ° C for 30 seconds and then with 4% pure acetone (100%) for 5 minutes. The slices were air dried and soaked in phosphate buffered saline for 10 minutes. Subsequently, all slices were incubated in phosphate buffered saline containing 0.3% hydrogen peroxide for 5 minutes to block endogenous peroxidase and then rinsed twice with phosphate buffered saline. The slices were then treated with porcine serum diluted 1:10 in normal and 4% bovine serum albumin phosphate and buffered saline for 5 minutes to block non-specific binding of antibodies.

In each case, the antibody was incubated in a humid atmosphere at room temperature for 30 minutes. The sections were first incubated with the monoclonal antibody C242 (C242: II) prepared in Example 1 in phosphate buffered saline containing 4% bovine serum albumin; followed by biofeeded horse anti-mouse immunoglobulin (Vestastain, manufactured by Vector Laboratories, Burlingame, CA, USA), diluted 1: 400 in phosphate buffered saline containing 4% bovine serum albumin, containing 2% porcine anti-rabbit immunoglobulin. and finally incubated with AvidinDH / Bionic Ice Radish Peroxidase H Complex (manufactured by Dakopatts A / S, Copenhagen, Denmark). After each incubation step, sections were soaked in phosphate buffered saline for 15 minutes. The sections were then treated for 15 minutes with the substrate, rinsed with phosphate buffered saline, stained with hematoxylin, and applied to glycerol gelatin (manufactured by Merck, Darmstadt, Germany). 10 mg of 3-amino-9-ethylcarbazole dissolved in 6 ml of dimethyl sulphoxide and diluted to 50 ml with 0.02 M sodium acetate solution, pH 5.5, containing 4 μΐ of 30% hydrogen peroxide Sigma, St. Louis, USA). The substrate solution was filtered before use. The sections were then examined. The results are reported in Table 1.

Table 1

Investigation of C242: II binding of colorectal tumor tissues and various normal tissues by staining with C242: II

Fabric Number of stained samples / total samples Colorectal tumor 26/41 Normal colon 8/16 milk glands 2/3 parotid 2/2 Skin 0/1 (Sweat glands are slightly stained) Liver 0/2 (bile ducts slightly stained) Kidney 0/1 Pancreas 2/2 (wires) Gastric 0/1 thin Belek 0/1

As can be seen from the data in Table 1, 63% of colorectal tumor tissue sections showed a positive result in the monoclonal C242: II binding assay. CA242 antigen expression was generally found to be lower in normal colon tissues than in tumorous tissues; staining was entirely confined to the colon epithelium and goblet cells. Non-colon normal tissues were essentially non-responsive to C242: II monoclonal antibody.

HU 215 243 Β

Although positive reactions were observed in normal mammary, pancreatic, bile, and sweat tissues, the staining intensity was low.

Immunohistochemistry of C242 (C242: II) in normal human tissue specimens after death

A large amount of normal tissue samples from a wide variety of tissues were isolated from a total of 21 dead organisms. For the C242: II immunoreactivity assay, 5 to 5 tissue samples were selected as the most responsive.

Immediately after death, isolated tissue samples were placed in a pre-cooled tube of antibiotic containing basic medium. Test tubes containing tissue samples were transported to the laboratory after storage in ice, media was removed, and tissue samples were cut into 0.5 cm ³ cubes. The tissue cubes were placed on strips of filter paper and frozen in liquid nitrogen. Frozen tissues were stored at -80 ° C until sectioning. For sectioning, 6 µm slices of frozen tissue samples were cut and adhered to a glass microscope slide. The slides were immersed in acetone at room temperature for 2 minutes to record slides. The slides were then removed from the acetone, air dried, and stored at -20 ° C until use.

In each case, the antibody was incubated in a humid atmosphere at room temperature for 30 minutes. The sections were first incubated with CBS42: II monoclonal antibody in TBS (Tris-buffered saline) and then with ice-cream diluted 1:50 in TBS containing 20% human serum (Sigma Chemical Company Ltd., Poole, Dorset, UK). peroxidase-conjugated rabbit anti-mouse IgG (Dako Ltd., High Wycombe, Buckinghamshire, United Kingdom) and finally diluted 1:50 in horseradish peroxidase-conjugated porcine anti-rabbit IgG (Dako Ltd., High Wycombe, Buckinghamshire, UK). Ltd.). After each incubation step, the sections were rinsed twice with TBS. The sections were then treated with substrate for 3-5 minutes, rinsed with TBS, stained with C242: II Mayers hematoxylin and applied to a synthetic support medium (Synthetic Mounting Medium by Shandon Scientific Ltd., Runcorn, Cheshire, UK). up.

The substrate used was 10 mg of diaminobenzidine (Sigma) in 17 ml of TBS containing 30% hydrogen peroxide. The substrate solution was filtered before use. The sections were then examined. The results are reported in Table 2.

Table 2

Examination of C242: II binding of normal tissues isolated from the dead by staining with C242: II

Fabric Number of stained samples / total samples Cerebellum 0/3 Brain 0/3 midbrain 0/3

Fabric Number of stained samples / total samples Heart (slight diffuse reactivity) Lungs 0/5 Peripheral nerves 0/3 Kidney (reactivity a 1/5 with proximal / distal tubes) Liver 1/5 (slight reactivity with conductors) Pancreas 2/2 (ductal and acinic) Colon 2/5 (mild reactivity with epithelial tissue) Small Intestine 5/5 (mild reactivity with epithelial tissue) Adrenal (slight reactivity with ^2/5 capsules) Bladder 1/5 (diffuse) Thyroid 1/5 (diffuse) Parathyroid 0/2 Skin 2/4 (scaly epithelial cells and sweat glands) Striated muscles 1/5 (diffuse) spleen 0/5 Lymph node 0/5 Gastric 5/5 (mild reactivity with epithelial cells) testicles 0/3 parotid 1/1 (slight reactivity with conductors) Almond (reactivity a 4/4 scaled epithelial cells)

As can be seen from Table 2, 72% of the tissue types tested did not react with C242: II. In the majority of the 28% of the responders, the staining was undefined and diffuse, or focused primarily on the tissue structure in the tissue. More heterogeneous but still minimal binding to gastrointestinal epithelial cells was observed. Two of the four skin tissue samples tested positive; reaction was observed on scaly epithelial cells and sweat glands. Each of the four almond tissue samples tested with C242: II showed a positive reaction on scaly epithelial cells.

Investigation of C242: II endocytosis of Colo 205 human colon cancer cells

Endocytosis was assayed as follows:

Individual cell suspensions were prepared from Colo 205 cells grown in monolayer culture (see above for the preparation of C242 antibody). The cells were trypsinized, washed extensively, and resuspended in culture medium. 5 ml of the suspension containing 10 ⁶ cells of the individual Colo 205 cells

HU 215 243 Β volumes were added to the tubes and ¹²⁵ I-labeled C242: II antibody (200,000 bp, corresponding to 50 ng monoclonal antibody) was added to the tubes. The final volume of the samples was 200 μl per tube.

Cells were incubated for 1 hour on ice (0 ° C) in the presence of ¹²⁵ I-C242. The suspension was then centrifuged and the cells were washed to remove unbound monoclonal antibody. The pre-incubated cells were further incubated for 10-10 minutes at 37 ° C; between each incubation step, the cell suspensions were cooled and the cells were pelleted. Supernatant radioactivity was measured using a LKB gamma counter (manufactured by Pharmacia AB, Uppsala, Sweden) to determine the amount of radiolabeled material released into the culture medium. Trichloroacetic acid was then added to the supernatant and the radioactivity of the precipitate was measured. Subsequently, ¹²⁵ I-C242 bound to the cell surface was removed by washing with 0.2 M glycine-HCl buffer (pH 1.5) containing 2.5 mg / ml papain, and the radioactivity of the cells was measured. The measured value indicates the amount of ¹²⁵ I-C242 internalized (incorporated in the cell). The amount of surface-bound antibody after each incubation step at 37 ° C was calculated by subtracting the amount of total surface-bound monoclonal antibody from the medium and the amount of internalized monoclonal antibody. The degradation of the antibody released into the medium was determined by measuring the radioactivity of the precipitate obtained by treatment of the supernatant with trichloroacetic acid.

The results are shown graphically in Figure 15. The figure shows the endocytosis of ¹²⁵ I-C242II in Colo 205 cells, the cell surface radioactivity ("Sur"), the internalized radioactivity ("Int"), the radioactivity released into the medium ("Slit") and the radioactivity of the antibodies released into the degraded medium. Deg.Re ”) as a percentage of the total initial radioactivity on the cell surface. Figure 15 shows that when preincubated cells are incubated for 1 hour at 37 ° C, more than 20% of C242: II is intimalized. Only a small amount of acid soluble radioactive material was observed after treatment of the supernatant with trichloroacetic acid, indicating that the antibody released into the culture medium was only slightly fragmented after 1 hour incubation.

Investigation of the immunotoxin cytotoxicity of ricin-A / C242 antibody

In vitro cytotoxicity assay

In this assay, the cytotoxicity of the immunotoxin against the human colorectal cell line (Colo 205 - ATCC No. CCL 222) was investigated.

Colo 205 cells were grown in RPM1640 medium containing 2.0 g / l sodium bicarbonate, glutamine free or 5% heat inactivated fetal bovine serum, mmol L-glutamine and 50 pg / ml gentamicin. Cells were counted using hemocytometer blocks. To investigate the inhibition of protein synthesis, 100 wells of 100-well (2 x 10 ⁴ cells / well) were added to 96-well plates. Samples of the immunotoxin were added in triplicate to each well. Immunotoxin was assayed in a 12-fold dilution series of double dilutions; the initial concentration was 2000 ng / ml and the final concentration was 0.98 ng / ml. In some wells, 100-100 μί medium was added as a control. Each plate was incubated for 24 hours at 37 ° C and then ³ H L-leucine with 2 pCi activity was added to each well. The plates were incubated for another 24 hours at 37 ° C. Subsequently, 50 µl trypsin was added to each well and the plates were incubated for an additional 15-20 minutes. The cells in each well were harvested into glass fiber webs and the radioactivity of the cells was determined using a scintillation counter (LKB's "Betaplate" counter). Protein synthesis inhibition curves were constructed from the measured values and IC ₅₀ values were determined from these curves. The results obtained with the most preferred immunotoxin (the preparation of which is described above) are below. 4A is shown.

Investigation of cytotoxicity in vivo

In this experiment, we investigated the effect of the immunotoxin on a human tumor cell line (Colo 205) grown as a subcutaneous foreign graft in mammary glandless mice. In the control studies, mice were given either antibody or phosphate buffered saline only.

5x10 ⁶ Colo 205 cells were injected into the flank of mice without thymus in a single subcutaneous injection. The tumors were allowed to grow and the size of the tumors was measured in two dimensions with a tactile compass every 3-4 days. The test substance was not administered until the tumors reached 0.7-1.0 cm ² . The test substance was then injected intravenously into the tail vein of the mice. Treatment on the first day of the experiment (day 0) was repeated on the first and second days of the experiment. Two-dimensional tumor size measurements were repeated every 3-4 days until the end of the experiment, and tumor reduction was calculated.

The study was performed on three groups of animals. Group 1 animals were dosed with 0.1 ml / 10 g phosphate buffered saline three times daily by intravenous injection. Group 2 animals were administered 1.2 mg / kg C242 antibody three times daily by intravenous injection. Group 3 animals were administered 2.0 mg / kg of the above immunotoxin three times daily by intravenous injection.

The results are shown in Table 3.

HU 215 243 Β

Table 3

Group- song Sun Relative size of tumors developed in each mouse First Second Third 4th 5th 6th 7th 8th 9th 10th First 0th 1 1 1 1 1 1 1 1 1 1 4th 2.02 1.15 2.28 1.02 0.51 0.80 2.05 1.58 0.79 1.23 8th 5.04 3.27 5.08 2.18 1.09 1.83 4.54 4.75 1.59 3.63 10th 6.41 4.32 5.24 3.94 1.96 2.44 5.58 5.91 1.85 4.43 15th 7.72 5.24 4.13 2.94 7.25 9.45 2.71 7.02 18th 11.81 6.78 3.53 8.02 14.39 3.30 11.48 Second 0th 1 1 1 1 1 1 1 1 1 1 4th 1.64 2.19 1.85 1.08 1.48 2.07 1.65 1.36 2.14 2.47 8th 3.31 4.30 3.28 4.07 3.38 4.78 6.08 2.96 3.34 7.31 10th 4.46 4.65 4.85 3.44 5.11 8.82 4.19 4.12 7.43 15th 5.29 10.80 2.57 18th 15.03 Third 0th 1 1 1 1 1 1 1 1 1 4th 0.63 0.86 1.40 0.64 0.48 0.43 0.73 0.91 0.77 8th 0.44 0.87 1.08 0.48 0.58 0.41 0.67 0.93 1.07 10th 0.48 0.71 1.39 0.29 0.55 0.51 0.56 1.13 1.21

As can be seen from the data in Table 3, the size of the tumors is reduced in immunotoxin-treated animals, unlike in phosphate-buffered saline only or antibody-only treated animals.

Reactivity of C242: II antibody in human pancreatic tumor tissues

Reactivity of C242: II antibody in human pancreatic tumor tissues was assayed using the same immunohistological method as described for reactivity in normal human tissues. Fresh pancreatic tumor tissue obtained by fine needle biopsy was frozen, and tissue samples from excised tumors were fixed with formalin and embedded in paraffin. Of the 16 pancreatic tumor tissue samples examined, 13 were stained positively. In general, binding to colorectal tumor tissues was found to be the same or better than that observed there.

To investigate the effect of immunotoxin on pancreatic tumor cell lines in vitro

The cytotoxicity of the immunotoxin was tested in vitro against a human pancreatic tumor cell line. Pan 1 cell line derived from pancreatic adenocarcinoma was used to quantify in vitro cytotoxicity and expression of the targeted antigen was confirmed by FACS analysis. To quantify the effect, the immunotoxin was added to varying concentrations of the Pan 1 cell culture and the IC _{50 was} determined. As negative controls, MOPC

21: Conjugate ricin-A and ricin-A were used.

The biological assay was performed in triplicate. In studies, the mean effective dose was 40 ± 18 ng / ml. Negative controls at 1000 ng / ml did not show cytotoxic activity either. These results demonstrate that the abdominal salivary gland tumor cells carry the targeted antigen and that the antigen-positive tumor cells are sensitive to the cell-killing activity of the immunotoxin.

Molecular cloning of cDNA coding portions of the light and heavy chains of the monoclonal antibody C242: II

A) Production of complete RNA

Total RNA was prepared from 10 ⁸ cells. Chomezynski P. and Sacchi N. [Anal. Biochem., 162, 156-159 (1987)] utilizes the Cirgwin method (Biochemistry, 18, 5294-5299 (1979)). Proceed as follows: The cell pellet was dissolved in 15 ml of cold denaturing solution containing 4 mol of guanidinium thiocyanate, 25 mmol of sodium citrate (pH 7), 0.5% N-lauroyl sarcosine and 0.1 mol of 2-mercaptoethanol. After dissolution of the cell material, the solution is 1.2 ml

A 2 molar sodium acetate solution (pH 4.0) was added and the mixture was extracted with phenol / chloroform / isoamyl alcohol (250: 49: 1). After centrifugation, RNA was precipitated from the aqueous phase with an equal volume of isopropanol. The mixture was kept at -20 ° C overnight to complete RNA precipitation. The RNA was separated by centrifugation, resuspended, the precipitation was repeated, and the RNA was again separated by centrifugation and the precipitation was repeated. Absorbance measurements show a total of 960 pg

RNA was obtained.

HU 215 243 Β

B) Isolation of mRNA

From the preparation of total RNA obtained above, the polyadenylated mRNA was isolated using a PolyATract (trademark) mRNA isolation system (manufactured by Promega Corporation; Madison, Wisconsin, USA) [see Promega Technical Bulletin 090 (1990)]. Proceed as follows: 960 pg of total RNA was dissolved in water and subjected to 50 picomoles of bio-oligo (dT) probe in 0.4 * SSC (1 * SSC = 8.77 g of sodium chloride and 4.41 g of sodium citrate) solution, pH 7.0). Streptavidin Magnesphere (streptavidin-coated) paramagnetic beads were added and after 10 minutes of incubation, the magnetic particles were separated and washed several times with 0.1 L SSC. Polyadenylated mRNA was resolved from the beads by aqueous incubation. 5 pg of mRNA was obtained.

C) Construction of a cDNA phage library in E. coli

The cDNA was prepared from the polyadenylated mRNA obtained in the previous step essentially according to the method of Gubler U. and Hoffman BJ (Gene, 25, 263-269 (1983)). The method was applied in a modified form to the Uni-ZAP (trademark; Stratagene Inc., La Jolla, CA, USA), which allows one-way cloning of the double-stranded cDNA. The procedure was as follows: 5 µg of polyadenylated mRNA was cleaved with 56 ng / ml Uni-ZAP coupling primer and 0.6-0.6 mmol dATP, dGTP, dTTP and 5-methyl-dCTP. and 45 units of Moloncy Murine Leukemia Virus reverse transcriptase was added to single-stranded cDNA. The second cDNA strand was synthesized by mixing 0.15-0.15 mmol of 1 mixture of dATP, dGTP, dTTP and dCTP, 3.2 units of RNase H and 6.5 units of DNA polymerase for 2.5 hours at 16 ° C. incubated. The mixture was extracted with phenol / chloroform (1: 1) and the cDNA precipitated with ethanol. The ends of the cDNA were blunted by incubation with 0.16 mmol dNTP and 10 units T4 DNA polymerase for 30 minutes at 37 ° C. The mixture was extracted with a mixture of phenol and chloroform, the cDNA was precipitated with ethanol, and the resulting blunt-ended cDNA was ligated to Eco RI adapters in the presence of 3 Weiss units of T4 DNA ligase and 1 mmol of ATP overnight at 8 ° C. incubated. After ligation, the cohesive ends of EcoRI were reacted with kinase by adding 1 mmol ATP and 10 units of T4 polynucleotide kinase and incubating for 30 minutes at 37 ° C. The resulting cDNA, which contains Eco RI phosphorylated at the cohesive end, was digested with 90 units of Xho I to form an Xho I cohesive sequence from the sequence encoded by the Uni-ZAP linker primer. The resulting cDNA was then ligated to 1 µg of Uni-ZAP XR EcoRI and Xhol in the presence of 2 Weiss units T4 DNA ligase and 1 mmol ATP; the mixture was incubated overnight at 12 ° C. The ligation mixture was introduced into in vitro phage particles using Gigapack II Gold (Trademark) extract, following the manufacturer's instructions, and the resulting phage were infected with E. coli strain PLK-F '. The resulting cDNA library of 8.5x10 ⁴ independent clones was scaled once to reach 7.5 x 10 ⁸ plaque-forming units / ml. The resulting cDNA library was screened using E. coli XL1Blue host cells (Bullock, W., Biotechniques, 5, 4 (1978)). The screening assay is described below.

D) Preparation of hybridization probes

For the detection of cDNA clones encoding the IgG1 heavy chain and kappa chain of antibody C242: II from mouse genomic DNA by the polymerase chain reaction of Saiki RH et al., Science 239, 487-491 (1988), the first constant region of the heavy chain (CH1) ) and hybridization probes covering the kappa chain constant region (CK) were prepared. Proceed as follows: Oligonucleotide primer pairs hybridizing to the 5 'and 3' regions of exons encoding the above immunoglobulin regions were used to amplify murine genomic DNA. The product was purified by agarose gel electrophoresis, and the resulting DNA probe fragments were randomly primed by Feinberg AP and Vogelstein B. Anal. Labeled Biochem., 132, 6 (1983)] ³² P version.

E) Screening and Plasmid Insertion of IgG1 Heavy and Kappa Coding Sequences

The cDNA library prepared under C. was screened in two separate steps with the immunoglobulin IgG1 and kappa probes prepared under B. Filtration was performed by the method of Sambrook J. et al., Molecular Cloning 2nd Edition, 1989 (Cold Spring Harbor Laboratory Press). The positively hybridizing phage clones obtained in the two separate hybridization assays were purified by re-filtration using the same probes. The positively hybridizing phage clones were propagated in E. coli CL1-Blue cultures and the resulting phage stock was used to construct cDNA-containing pBluescript SK (-1) essentially according to the method of Short JM et al., Nucleic Acids Res., 16, 7583-7600. (1988)].

F) Determination of plasmid cDNA characteristics from hybridization colonies

Characteristics of the resulting cDNA-containing plasmids were determined by restriction enzyme mapping, and plasmids containing the expected size insert were analyzed by Sanger.

F. et al., Proc. Natl. Acad. Sci., USA, 74, 5463-5467 (1977)] was subjected to dideoxy DNA sequencing. One plasmid hybridizing with the IgK kappa probe, pKGE761, contained a cDNA insert having the sequence of known murine kappa light chain sequences. Health and Human Services, National Institutes of Health)] encodes a near-homologous open-reading frame. The cDNA partial sequence encoding the light kappa chain variable region (VK) and its corresponding protein sequence are shown in Figure 18. The three CDRs CDR1, CDR2 and CDR3 are underlined28

HU 215 243 Β. The signal peptide before the amino-terminal end of the VK segment was designated S. The constant region following the carboxy-terminal end of the VK sequence is designated CK.

One plasmid hybridizing with an IgG1 probe, pKGE762, contained a cDNA insert having a sequence encoding a homologous open reading frame close to the known murine IgG1 heavy chain sequences (see Kabat et al., Cited above). The cDNA portion sequence encoding the IgG1 heavy chain variable region (VH) and its corresponding protein sequence are shown in Figure 19. The three CDRs CDR1, CDR2 and CDR3 are underlined. The signal peptide before the amino-terminal end of the VH segment is designated S. The constant region following the carboxy-terminal end of the VH sequence is designated CH1.

PREPARATION OF THE MEDICINAL PRODUCT An injectable sterile aqueous solution of the following composition for therapeutic use in human medicine was prepared by known pharmaceutical technology:

mg / ml solution

Ricin-A / C242 antibody immunotoxin 1.0

Sodium acetate trihydrate 6.8

Sodium chloride 7.2

Tween 20 0.05

List of sequences

SEQ ID NO. 1:

The sequence features:

(a) length: 16 amino acids (b) type: amino acid (c) fiber type: single-stranded (d) topology: linear

Sequence description:

Arg Sex Ser Lys Ser Leu Leu His Leu Tyr SEQ ID NO. 2: Ser Asn Gly Asn Thr Tyr Type (b): amino acid (c) fiber type: single-stranded (d) topology: linear 14 16 7 The sequence features: (a) Length: 7 amino acids Sequence description: With Ser 25 Arg Met Ser Asn Leu SEQ ID NO. 3: Sequence Characteristics: (a) Length: 9 amino acids 30 Type (b): amino acid (c) fiber type: single-stranded (d) topology: linear Sequence description: Leu Gin His Leu Glu Tyr Pro Phe Thr 9

SEQ ID No. 4: Type (b): amino acid

Sequence characteristics: (c) fiber type: single-stranded (a) length: 5 amino acids (d) topology: linear

Sequence description:

Tyr Thr Gly Met Asn 5 SEQ ID NO. 5: Sequence characteristics: (a) Length: 17 amino acids Type (b): amino acid (c) fiber type: single-stranded (d) topology: linear Sequence description: 45 Trp Ile asp Thr Thr Thr Gly Glu Pro Thr Tyr Alá Glu Asp 14 Phe Lys Gly 17 SEQ ID NO. 6: Sequence Characteristics: (a) Length: 10 amino acids 50 Type (b): amino acid (c) fiber type: single-stranded (d) topology: linear Sequence description: Arg Gly Pro Tyr Asn Trp Tyr Phe Asp Val 10 SEQ ID NO. 7: Sequence Characteristics: (a) Length: 18 amino acids Type (b): amino acid (c) fiber type: single-stranded (d) topology: linear

HU 215 243 Β

Sequence description:

Arg Ser Ser lys Ser Leu Leu Tyr Trp Phe

SEQ ID NO. 8:

The sequence features:

(a) Length: 11 amino acids

Sequence description:

Ile Tyr Arg Met Ser Asn

SEQ ID NO. 9:

The sequence features:

(a) Length: 11 amino acids

Sequence description:

Leu Gin His Leu Glu Tyr

SEQ ID NO. 10:

The sequence features:

(a) Length: 7 amino acids

Sequence description:

Phe Thr Tyr Thr Gly Met

SEQ ID NO. 11:

The sequence features:

(a) Length: 21 amino acids

Sequence description:

Met Gly Trp Ile Asp Tha Glu Asp Phe Lys Gly Arj

Leu His Ser Asn Gly Asn Thr Tyr 14 18 (b): amino acid (c) fiber type: single-stranded (d) topology: linear

Leu Val Ser Gly Val Type 11 (b): amino acid (c) fiber type: single-stranded (d) topology: linear

Pro Phe Thr Phe Gly 11 (b) Type: Amino Acid (c) Fiber Type: Single Thread (d) Topology: Linear

Asn 7 (b): amino acid 25 (c) fiber type: single-stranded (d) topology: linear ¹ Thr Thr Gly Glu Pro Thr Tyr Ala 14: Ile 21

SEQ ID NO. 12:

The sequence features:

(a) Length: 14 amino acids

Sequence description:

Ala Arg Arg Gly Pro Tyr (b) type: amino acid (c) fiber type: single-stranded 35 (d) topology: linear

Asn Trp Tyr Phe Asp Val Trp Gly 14

SEQ ID NO. 13:

The sequence features:

(a) Length: 25 bases

Sequence description:

AATTCGCAT GCGGATCCAT CGATC

SEQ ID NO. 14:

The sequence features:

(a) Length: 25 bases

Sequence description:

GCGTACGCC TAGGTAGCTA GAGCC

SEQ ID NO. 15:

The sequence features:

(a) Length: 21 bases

Sequence description:

GATAACAACA TATTCCCCAA Type A (b): nucleic acid (c) fiber type: single stranded (d) topology: linear (b) type: nucleic acid (c) fiber type: single stranded (d) topology: linear (b) type: nucleic acid (c) fiber type: single-threaded 55 (d) topology: linear

HU 215 243 Β

SEQ ID NO. 16:

The sequence features:

(a) Length: 23 bases

Sequence description:

GATAACAAC ATGGTACCC AAA

SEQ ID No. 17:

The sequence features:

(a) Length: 23 bases

Sequence description:

AACAACATG GTACCCAAA CAA

SEQ ID NO. 18:

The sequence features:

(a) Length: 1140 bases

Sequence description:

TTCTGGCAAA TATTCTGAAA

TTAACTAGTA CGCAAGTTCA

GGATCCACCT CAGGGTGGTC

CAAACAATAC CCAATTATAA

GCTACACAAA CTTTATCAGA

GATGTGAGAC ATGAAATACC

AAACCAACGG TTTATTTTAG

TTACATTAGC CCTGGATGTC

GGAAATAGCG CATATTTCTT

AATCACTCAT CTTTTCACTG

GTGGTAATTA TGATAGACTT

ATCGAGTTGG GAAATGGTCC

TTACAGTACT GGTGGCACTC

TTTGCATCCA AATGATTTCA

GAAATGCGCA CGAGAATTAG

CGTAAMACA CTTGAGAATA

AGTCTAACCA AGGAGCCTTT

GGTTCCAAAT TCAGTGTGTA

TCTCATGGTG TATAGATGCT

TTATAAGGCC AGTGGTACCC

GCAAGCTTAG CCCGCCTAAT

TTGAGAGCCT TCAACCCAGT

TATCGTCGCC GCACTTATGA

SEQ ID NO. 19:

The sequence features:

(a) Length: 26 bases Sequence description:

ATAACAACAT

GGTTCCCAAA CAATAC (b) type: nucleic acid (c) fiber type: single stranded (d) topology: linear (b) type: nucleic acid (c) fiber type: single stranded (d) topology: linear (b) type: nucleic acid 15 (c) fiber type: single-threaded (d) topology: linear

TGAGCTGTTG ACAATTAATC ATCGAACTAG 50 CGTAAAAAGG GTATCGACAA TGGIACCCGG 100 TTTCACATTA GAGGATAACA ACATGGTACC 150 ACTTTACCAC AGCGGGTGCC ACTGTGCAAA 200 GCTGTTCGCG GTCGMTAAC AACTGGAGCT 250 AGTGTTGCCA AACAGAGTTG GTTTGCCTAT 300 TTGAACTCTC AAATCATGCA GAGCTTTCTG 350 ACCAATGCAT ATGTGGTCGG CTACCGTGCT 400 TCATCCTGAC AATCAGGAAG ATGCAGAAGC 450 ATGTTCAAAA TCGATATACA TTCGCCTTTG 500 GAACAACMG CTGGTAATCT GAGAGAAAAT 550 ACTAGAGGAG GCTATCTCAG CG1BTTTATTA 600 AGCTTCCAAC TCTGGCTCGT TCCTTTATAA 650 GAAGCAGCAA GATTCCAATA TATTGAGGGA 700 GTACAACCGG AGATCTGCAC CAGATCCTAG 750 GTTGGGGGAG ACTMCCACT GCAATTCAAG 800 GCTAGTCCAA TTCAACTGCA AAGACGTAAT 850 CGATGTGAGT ATAMAATCC CTATCATAGC 900 CACCTCCACC ATCGTCACAG TTMGATTGC 950 GGGGATCCTC TAGAGTCGAC CTGCAGGCAT 1000 GAGCGGGCM TTTTTTATCG ACCGATGCCC 1050 CAGCTCCTTC CGGTGGGCGC GGGGCATGAC 1100 CTGTCTTCTT TATCATGCAA 1140

(b) type: nucleic acid (c) fiber type: single-stranded (d) topology: linear

HU 215 243 Β

SEQ ID NO. 20:

The sequence features:

(a) Length: 86 bases

Sequence description:

AAAAAGGGTA TCGACATGGT (b) type: nucleic acid (c) fiber type: single-stranded (d) topology: linear

AGCCGGGGAT CCACCTCAGG GTGGTCTTTC '50

ACATTAGAGG ATAACAACAT GGTACCCAAA CAATAC 86

SEQ ID No. 21: Type (b): nucleic acid

Sequence Characteristics: (c) Fiber Type: Single Thread (a) Length: 463 Base 10 (d) Topology: Linear

Sequence description:

GGAATTCGGC ACGAGGAGTT TTTTTGTACT AAGTTCTCAGA ATG AGG TGC 49 Met Arg Cys

CTA GCT GAG TTC CTG GGG CTG CTT GTG CTC TGG ATC CCT GGA 91 Leu Ala Glu Phe Leu Gly Leu Leu With Leu Trp Ile Pro Gly GCC ATT GGG DAM ATT GTG ATG ACT CAG GCT GCA CCC TCT GTA 133 Ala Ile Gly asp Ile With Met Thr Gin Ala Ala Pro Ser With CCT GTC ACT CCT GGA GAG TCA GTA TCC ATC TCC TGC AGG TCT 175 Pro With Thr Pro Gly Glu Ser With Ser Ile Ser Cys Arg Ser AGT AAG AGT GTC CTG CAT AGT AAT GGC AAC ACT TAC TTG TAT 217 Ser Lys Ser Leu Leu His Ser Asn Gly Asn Thr Tyr Leu Tyr TGG TTC CTG CAG AGG CCA GGC CAG TCT CCT CAG CTC CTG OVER THE 259 Trp Phe Leu Gin Arg Pro Gly Gin Ser Pro Gin Leu Leu Ile TAT CGG ATG TCC AAC CTT GTC TCA GGA GTC CCA GAC AGG TTC 301 Tyr Arg Met Ser Asn Leu With Ser Gly With Pro asp Arg Phe AGT GGC AGT GGG TCA GGA ACT GCT TTC ACA CTG AGA ATC AGT 343 Ser Gly Ser Gly Ser Gly Thr Ala Phe Thr Leu Arg Ile Ser AGA GTG GAG GCT GAG DAM GTG GGT GTT TAT TAC TGT CTG CAA 385 Arg With Glu Ala Glu asp With Gly With Tyr Tyr Cys Leu Gin CAT CTA GAG TAT CCG TTC ACG TTC GGT CCT GGG ACC AAG CTG 427 His Leu Glu Tyr Pro Phe Thr Phe Gly Pro Gly Thr Lys Leu GAG CTG AAA CGG GCT DAM GCT GCA CCA ACT GTA ACG 463 Glu Leu Lys Arg Ala asp Ala Ala Pro Thr With Thr

SEQ ID NO. 22: Type 45 (b): nucleic acid The sequence features: (c) fiber type: single-stranded (a) Length: 483 bases (d) topology: linear Sequence description:

GAATTCGGCA CGAGATTGAG CCCAAGTCTT AGACATC ATG GAT TGG CTG CGG 52 Met Asp Trp Leu Arg

AAC TTG CTA TTC CTG ATG GCA GCT GCC CAA AGT ATC CAA GCA CAG 97 Asn Leu Leu Phe Leu Met Ala Ala Ala Gin Ser Ile Gin Ala Gin GTC CAG TTG GTG CAG TCT GGA CCT GAG CTG AAG AAG CCT GGA GAG 142 With Gin Leu With Gin Ser Gly Pro Glu Leu Lys Lys Pro Gly Glu ACA GTC AAG ATC TCC TGC AAG GCT TCT DAM TAT AOC TTC ACA TAC 187 Thr With Lys Ile Ser Cys Lys Ala Ser asp Tyr Thr Phe Thr Tyr

HU 215 243 Β

TAT GGA ATG AAC TGG GTG AAG CAG GCT CCG GGA AAG GGT TTA AAG 232 Tyr Gly Met Asn Trp Val lys Gin below Pro Gly Lys Gly Leu Lys TGG ATG GGC TGG OVER THE GAC ACC ACC ACT GGA GAG CCA ACA TAT GCT 277 tip llet Gly Trp Ile asp Thr Thr Thr Gly Glu Pro Thr Tyr below GAA DAM TTT AAG GGA CGG ATT GCC TTC TCT TTG GAG ACC TCT GCC 322 Glu asp Phe lys Gly Arg Ile below Phe Ser Leu Glu Thr Ser below AGC ACT GCC TAT TTG CAG ATC AAA AAC CTC AAA AAT GAG GAC ACG 367 Ser Thr below Tyr Leu Gin Ile Lys Asn Leu Lys Asn Glu asp Thr GCT ACA TAT TTC TGT GCA AGA CGG GGG CCT TAC AAC TGG TAC TTT 412 below Thr Tyr Phe Cys below Arg Arg Gly Pro Tyr Asn Trp Tyr Phe DAM GTC TGG GGC GCA GGG ACC ACG GTC ACC GTC TCC TCA GCC AAA 457 asp With Trp Gly below Gly Thr Thr With Thr With Ser Ser below Lys ACT AON CCC CCA TCT GTC TAT CC 483 Thr Thr Pro Pro Ser With Tyr Pro

PATENT CLAIMS

Claims (15)

PATENT CLAIMS A process for the preparation of a conjugate comprising a toxin component and a target cell binding component selectively bound to gastric and intestinal tumor cells, namely a C242 antibody or a target cell binding component having substantially the same cell binding properties as the C242 antibody, characterized in that reacting the toxin component with the target cell binding component to produce a conjugate comprising directly linking the target cell binding component, or with a coupling agent derivative of the target cell binding component. The method of claim 1, wherein the target cell binding component is C242 antibody or fragment thereof. The process according to claim 1 or 2, wherein the toxin component is recombinant ricin A. 4. A method according to any one of claims 1 to 6, wherein the target cell binding component is coupled to the toxin component via a bifunctional coupling agent. The method according to any one of the preceding claims, wherein the target cell binding component is coupled to the toxin component via a disulfide or thioether linkage. The method of any one of the preceding claims, wherein the target cell-binding component is linked to the toxin component via a -SX-CO- linking moiety, wherein X is optionally one or two C 1 -C 6 alkyl, phenyl or -C 1-6 straight-chain alkylene substituted with C 4 alkyl) phenyl, or optionally substituted on the alkyl group with one or two C 1-6 alkyl, phenyl or C 1-4 alkylphenyl groups and optionally halogen on the benzene ring, C 1-4 alkyl or phenoxy substituted with C 1-4 alkoxy or C 1-4 alkoxy. 7. The method of claim 6, wherein the target cell binding component is coupled to the toxin component via a -SX-CO- linker moiety wherein X is methylene, ethyl, propyl or butyl substituted with one or two methyl, ethyl, propyl or butyl groups. represents an ethylene, propylene or butylene group. The method of claim 6, wherein the target cell binding component is coupled to the toxin component via a -S-CH (CH ₃ ) CH ₂ -CO- linker moiety. The method of any one of the preceding claims, wherein the target cell binding component is a light chain having substantially the following sequence of CDRs: CDR1: ArgSerSerLysSerLeuLeuHisSerAsnGlyAsnThrTyrLeuTyr CDR2: ArgMetSerAsnLeuValSer CDR3: LeuGlnHisLeuGluTyrProPheThr heavy chain: CDR1: TyrTyrGlyMetAsn CDR2: TrpIleAspThrThrThrGlyGluProThrTyrAlaGluAspPheLysGly CDR3: ArgGlyProTyrAsnTrpTyrPheAspVal uses a component containing an arrangement that provides cell binding properties substantially equivalent to those of the C242 antibody. The method of claim 1 For the preparation of a conjugate of the formula -S, -S ₂ -X-CO-NH-B in the formula HU 215 243 Β Recombinant means ricin A, B represents the target cell binding component, namely C242 antibody, S / a represents the sulfur atom in recombinant ricin A, the A / f group is on C242, S represents ₂ sulfur atoms, and X is a straight-chained C 1-6 alkylene group optionally substituted with one or two C 1-6 alkyl, phenyl or (C 1-4 alkyl) phenyl or optionally substituted on the alkyl group with one or two C 1-6 alkyl, phenyl - or (C1-C4-alkyl) -phenylene substituted with (C1-C4) -alkyl-phenyl and optionally substituted on the benzene ring by a halogen, (C1-C4) -alkyl or (C1-C4) -alkoxy-, characterized in that the target cell binding moiety B-NH-CO-X ^ Lj derivative of general formula: - wherein Β, X and S ₂ are as stated, L is a leaving group - is reacted with a recombinant ricin-A. 11. The method of claim 1 for the preparation of a conjugate comprising a C242 antibody and a recombinant ricin-A, wherein an amino group of the C242 antibody is linked via a -CO-CH ₂ -CH (CH ₃ ) -S- derivative from 3-mercaptobutyric acid. to a thiol group in a cysteine residue of recombinant ricin A, characterized in that C242antitest -NH-CO-CH ₂ CH _(i) -SL derivative general formula _I - wherein the NH is on the C242 antibody and t represents pyridyl ./- - presence of ricin-A. The method of claim 1, wherein said recombinant ricin A is a toxin component and at least one of the CDRs having substantially the following sequence is a light chain: CDR1: ArgSerSerLysSerLeuLeuHisSerAsnGlyAsnThrTyrLeuTyr CDR2: ArgMetSerAsnLeuValSer CDR3: LeuGlnHisLeuGluTyrProPheThr heavy chain: CDR1: TyrTyrGlyMetAsn CDR2: TrpIleAspThrThrThrGlyGluProThrTyrAlaGluAspPheLysGly CDR3: ArgGlyProTyrAsnTrpTyrPheAspVal for the preparation of a conjugate comprising a target cell binding component having an array of a target cell binding component having an arrangement of the component having essentially the same cell binding properties of the C242 antibody, characterized in that or (b) coupling the toxin component to the target cell binding component via a coupling agent to form a conjugate of the target cell binding component with a toxin component coupling derivative or a toxin component binding to a target cell binding component. 13. A process for preparing a conjugate comprising a toxin component and a target cell binding component specific for a gastrointestinal tumor epitope, comprising (a) reacting a target cell binding a toxin component to a conjugate comprising a toxin component and a target cell binding component; or (b) reacting the target cell binding component with a coupling derivative of the toxin component or the coupling derivative of the target cell binding component to produce a conjugate comprising the toxin component and the target cell binding component via a coupling agent. 14. The method of claim 13, wherein said C242 antibody-binding epitope-specific cell-binding component is used. 15. A process for the preparation of a pharmaceutical composition containing the conjugates as an active ingredient, wherein the conjugates prepared according to any one of the preceding claims are converted into a pharmaceutical composition using conventional pharmaceutical diluents, carriers and / or other excipients.